Citation
Partners of men with spinal cord injury

Material Information

Title:
Partners of men with spinal cord injury an exploration of stress, depression and burden
Creator:
Charlifue, Susan B
Publication Date:
Language:
English
Physical Description:
xiv, 191 leaves : ; 28 cm

Subjects

Subjects / Keywords:
Spinal cord -- Wounds and injuries -- Patients -- Family relationships ( lcsh )
Spouses ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 180-191).
Statement of Responsibility:
by Susan B. Charlifue.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
57494825 ( OCLC )
ocm57494825
Classification:
LD1190.L566 2004d C42 ( lcc )

Downloads

This item has the following downloads:


Full Text
PARTNERS OF MEN WITH SPINAL CORD INJURY:
AN EXPLORATION OF STRESS, DEPRESSION AND BURDEN
by
Susan B. Charlifue
B.A., University of Denver, 1976
M.A., University of Denver, 1986
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Doctor of Philosophy
Health and Behavioral Sciences
2004


2004 by Susan B. Charlifue
All rights reserved.


This thesis for the Doctor of Philosophy
degree by
Susan B. Charlifue
has been approved
by
David P. Tracer
Deborah S. Main
Date
Gale G. Whiteneck


Charlifue, Susan B. (Ph.D., Health and Behavioral Sciences)
Partners of Men with Spinal Cord Injury: An Exploration of Stress, Depression and
Burden
Thesis directed by Associate Professor David P. Tracer
ABSTRACT
A spinal cord injury (SCI) can have a devastating affect on both the individual
sustaining the injury and family members. Within an overarching conceptual
framework of attachment theory, a sample of 137 couples who were married before
the SCI and had remained married a minimum of 5 years post-injury were interviewed
to explore quantitative and qualitative perceptions of stress, depression and burden in
wives of men with SCI. Stress was measured by the Perceived Stress Scale (PSS),
depression by the Center for Epidemiological Studies Depression Scale (CES-D), and
burden by the Caregiver Burden Inventory (CBI). Nine couples participated in
additional qualitative interviews. The study hypotheses were: 1) partners of men
with SCI will report greater stress and depression than their injured partners; 2)
partners of men with SCI who provide more than 10 hours per week of assistance will
report greater stress, depression and burden than those providing 10 hour or less of
assistance; 3) older partners of men with SCI will report less stress, depression and
burden than younger partners; and 4) in couples who demonstrate greater similarity of
attachment styles, female partners will report less stress, depression and burden than
in those couples with less similar attachment styles.
Results indicate that the men with SCI were significantly more depressed than their
wives. Wives who provided more than 10 hours of assistance had significantly
greater perceived burden than those who provided 10 or fewer hours of assistance.
Older spouses had significantly higher PSS scores than younger spouses. The only
significant difference related to attachment style was in the Emotional Burden
subscale of the CBI, which was higher for those in relationships with more similar
attachment styles. The qualitative interviews highlighted the importance of
establishing independence for husbands and wives, and the importance of family and
of social support on helping minimize perceptions of stress, depression and burden.
IV


This investigation has provided new and unique insights into the nature of marriages
that survive SCI, offering information to health and service providers to guide them in
developing intervention programs that may help families cope more effectively with
the consequences of SCI.
This abstract accurately represents the content of the candidates thesis. I recommend
its publication.
Signed
David P. Tracer
v


DEDICATION
I dedicate this thesis first and foremost to my husband, Gerry, for his support,
encouragement, and patience. He willingly took on additional responsibilities within
our household that I, too often, neglected while completing this degree. His love and
faith in me never flagged during this process, for which I am extremely grateful.
I also dedicate this to my parents, Miriam and Ralph Wexler, who instilled in me a
spirit of independence, curiosity, and desire to succeed. Their confidence in my
abilities has served as a gentle driving force throughout my life.
Finally, I dedicate this to those individuals with spinal cord injuries and their partners
who shared their stories with me. They have taught me how resolute the human spirit
can be, and how adversity can be addressed, acknowledged, and overcome. I have
drawn strength from their strength.


ACKNOWLEDGMENTS
I would like to acknowledge the outstanding support and help from my thesis
committee chair, Dr. David P. Tracer, and from my committee members, Dr. Lucinda
Bryant, Dr. Ann Gonzales, Dr. Deborah Main, and Dr. Gale Whiteneck. They each
provided expertise and insight, as well as continued encouragement in helping me
prepare this thesis.
1 would also like to acknowledge my colleagues in the Research Department of Craig
Hospital, particularly Kenneth Gerhart, MS, RPT; Hang Pham-Bowman, MA, Kendra
Noble, BA, Chris Cusick, MA, Cynthia Harrison-Felix, PhD, and Amy Dannels, ND,
RN. All were instrumental in providing support, helping with the data collection,
providing insight and suggestions to facilitate the data analysis, and generally acting
as a cheering squad during this process.
This project was supported by grant number H133000001 from the National Institute
on Disability and Rehabilitation Research, U.S. Department of Education Office of
Special Education and Rehabilitative Services.


CONTENTS
Figures.......................................................xii
Tables.......................................................xiii
CHAPTER
1. INTRODUCTION...................................................1
Rationale for Conducting the Study.........................2
Arrangement of the Thesis..................................4
2. REVIEW OF THE LITERATURE.......................................6
Introduction...............................................6
The Nature of Caregiving...................................8
Caregiver Studies with Non-SCI Samples....................14
Burden................................................14
Stress................................................18
Caregiver Health and Depression.......................26
Caregiver Studies with SCI Samples........................30
Attachment Theory.........................................36
Qualitative Aspects of Caregiving.........................40
viii
Specific Aims
48


Hypotheses....................................................48
3. STUDY DESIGN.......................................................50
Participants................................................. 50
Enrollment.....................................................52
Power Analysis.................................................54
Qualitative Sample.............................................55
Instruments....................................................56
Center for Epidemiological Studies Depression Scale (CES-D). 57
Perceived Stress Scale (PSS)...............................58
Caregiver Burden Inventory (CBI)...........................58
Revised Adult Attachment Scale (RAAS)......................59
Personal Support/Assistance Questionnaire..................60
Life Story Interviews......................................61
Analyses.......................................................62
Quantitative Analysis......................................62
Qualitative Analysis.......................................67
4. QUANTITATIVE RESULTS...............................................70
Sample Bias Analyses...........................................70
Caregiver Study Sample Characteristics.........................72
ix


Tests of Study Hypotheses
73
Hypothesis 1: Partners of men with SCI will report
greater a) stress; b) depression than their injured partners.73
Hypothesis 2: Partners of men with SCI who provide
more than 10 hours per week of assistance will report
greater a) stress; b) depression; c) burden than those
providing 10 hours or less of assistance.....................74
Hypothesis 3: Older partners of men with SCI will
report less a) stress; b) depression; c) burden than
younger partners.............................................77
Hypothesis 4: In couples where greater similarity
of attachment styles is demonstrated, female partners
will report less a) stress; b) depression; c) burden than
those in couples with less similar attachment styles.........79
Regression Analyses...............................................82
5. QUALITATIVE RESULTS...................................................92
Personal Support/Assistance Questionnaire.........................92
Life Story Interviews.............................................97
General Themes...............................................99
Critical Events.....!.....................................109
Significant People..........................................120
Future Script...............................................124
Stresses and Problems.......................................127
Personal Ideology...........................................129
x


6. DISCUSSION
141
7. CONCLUSIONS..................................152
APPENDIX
A. DATA COLLECTION INSTRUMENTS..............157
B. IRB APPROVAL LETTERS....................173
REFERENCES...................................... 180
xi


FIGURES
Figure
1 Alzheimers Caregiver Stress Model...................................19
2 Stress Appraisal Model...............................................22
3 Common Characters in Personal Myth...................................45
4 Proposed SCI Caregiver Stress Model.................................149
xii


TABLES
Table
1 Summary of Relevant Literature Non-SCI Studies.....................11
2 Summary of Relevant Literature SCI Studies.........................13
3 Potential Eligible Candidates.......................................53
4 Description of Linear Regression Analyses...........................66
5 Comparison of Characteristics of Individuals Included
and Not Included in the Study.......................................71
6 Demographic Characteristics of Female Spouses in the Study..........72
7 Comparison of PSS and CES-D Scores Between Men with SCI
and their Partners..................................................74
8 Bivariate Regression Analysis of Hours Per Week Caregiving
with CES-D, PSS and CBI Scores......................................75
9 Comparison of PSS, CES-D and CBI Scores by Hours Per Week
of Assistance Provided to Spouse with SCI...........................76
10 Bivariate Regression Analysis of Female Spouses Age
with CES-D, PSS and CBI Scores......................................77
11 Comparison of PSS, CES-D and CBI Scores by Two Age Groups..........78
12 Attachment Styles of Individuals and Couples.......................80
13 Comparison of PSS, CES-D and CBI Scores Among Female
Partners by Similarity of Attachment Style..........................81
xiii


14 Bivariate Pearson Correlation Coefficients for Selected Predictor
and Outcome Variables..............................................83
15 Regression Analysis for Female Spouses CES-D Scores.............84
16 Regression Analysis for Female Spouses PSS Scores................85
17 Regression Analysis for CBI Total Scores.........................86
18 Regression Analysis for CBI Time Dependence Burden...............87
19 Regression Analysis for CBI Developmental Burden.................88
20 Regression Analysis for CBI Physical Burden......................89
21 Regression Analysis for CBI Social Burden........................89
22 Regression Analysis for CBI Emotional Burden.................:....90
23 Themes Regarding What Wives Would Have Changed....................93
24 Description of Qualitative Sample.................................98
xiv


CHAPTER 1
INTRODUCTION
Serious physical trauma or illness, while undoubtedly devastating to the individual
incurring such an event, can also have a far-reaching impact on family members and
close friends. This is particularly true when the trauma results in limitations that
necessitate physical assistance from others. Spinal cord injury (SCI) is one such
trauma that often results in dependence of the injured party on other people. As such,
it provides an opportunity to evaluate the broader consequences of disability on
family and friends; in particular, the impact of such a disability on a marital
relationship. Given the young age at which most of these injuries occur, it is not
surprising that more than half of individuals are single at the time of injury, and
approximately 31% are married (the remainder are divorced, separated or widowed at
the time of injury) (National Spinal Cord Injury Statistical Center, 2003). Divorce
following SCI is relatively frequent, with a rate at least as high as the nearly 50%
noted in the general population (DeVivo & Fine, 1985; Kreuter, 2000). The
percentage of individuals divorced continues to rise in the years following injury
(National Spinal Cord Injury Statistical Center, 2003). In the general population,
evidence suggests married individuals have lower risk of mortality than those who are
divorced, widowed or never married (Smith & Zick, 1994; Johnson, Backlund, Sorlie,


& Loveless, 2000; Kravdal, 2001; Lin, Rogot, Johnson, Sorlie, & Arias, 2003). Being
married also results in a lower risk of various poor health outcomes, including
depression (Caimey, Boyle, Offord, & Racine, 2003), injury (Hokby, Reimers, &
Laflamme, 2003), obesity and cardiovascular disease (Hedblad et al., 2002). The
beneficial impact of marriage on life satisfaction for people with SCI also has been
demonstrated (Post, de Witte, van Asbeck, van Dijk, & Schrijvers, 1998; Holicky &
Charlifue, 1999; Dowler, Richards, Putzke, Gordon, & Tate, 2001). Therefore, it can
be reasonably postulated that maintaining marriages post-injury may contribute to
long-term well-being for the person with SCI. A question remains, though, whether
the spouse shares this benefit. A spouse who has a partner with SCI may instead
experience depression or stress related to the injury and its other negative sequellae.
A spouse who provides personal assistance may feel some sense of burden as well.
However, it is also possible that a spouse may experience a sense of fulfillment,
usefulness or other positive benefits of assuming a caregiving role for a partner with a
severely disabling condition. The long-term consequences of SCI on the partner,
therefore, warrant greater investigation.
Rationale for Conducting the Study
Because SCI usually occurs in young male adults, this population provides an ideal
group among whom to study the impact of long-term caregiving within marriage. A
2


significant body of research, which is described in detail in Chapter 2, indicates that
caregiving spouses have high risk for stress, depression, experiencing a sense of
burden, and even physical deterioration. While this existing research is plentiful, very
little focuses on spouses of individuals with SCI, and almost none addresses these
issues in the years following the injury. Therefore, this study explores perceptions of
stress, depression and burden in female spouses of men with SCI who were married
prior to the injury and have remained married for at least 5 years post-injury.
Attachment theory, a conceptual framework that describes how individuals make
strong affectional bonds to others (Bowlby, 1977), provides a theoretical basis within
which to examine the nature of the relationship between partners to determine
whether or not similarity in attachment style affects caregiver stress, depression or
burden.
This study explores the impact of SCI on female spouses, whether or not they provide
personal assistance to their injured partners. More specifically, this study helps
address the following questions:
1. Do wives of men with SCI report greater stress and depression than their
injured partners?
3


2. Do wives of men with SCI who provide more than 10 hours of assistance per
week of assistance report greater stress, depression and burden than those
providing 10 hours or less of assistance?
3. Do older wives of men with SCI report less stress, depression and burden than
younger wives?
4. Does greater congruence in attachment in couples contribute to less reported
stress, depression and burden by wives?
The findings from this study are important for two reasons: 1) they will remediate a
large gap in the literature which, to date, has not focused on the long-term
consequences of caregiving in SCI, and 2) they will help guide rehabilitation
professionals who work with couples following SCI to help identify potential areas of
risk for marital discord and build on marital strengths early in the post-injury phase.
/
Arrangement of the Thesis
This thesis comprises seven chapters. Chapter One provides a brief background and
rationale for conducting this study. Chapter Two reviews the theoretical and
empirical background for this study, describing the known physical consequences of
spinal cord injury, the emotional impact of caregiving, and providing the background
of attachment theory, which guided this study.
4


Chapter Three describes the study design and methodology, including descriptions of
the instrumentation, sampling and analysis procedures conducted. Chapter Four gives
the demographic details of the sample and results of quantitative analyses of the
collected data. Chapter Five reports findings from qualitative analyses of interviews
conducted with a subsample of couples and discusses the themes and patterns that
emerged from the interviews.
Chapter Six provides an overview and discussion of the quantitative findings,
enhanced by information from the qualitative data, and compares these findings with
the existing research, and describes the limitations of the study. The thesis concludes
with Chapter Seven, which discusses the potential clinical implications of these
findings and directions for future research.
5


CHAPTER 2
REVIEW OF THE LITERATURE
I am his lover, his wife, his friend, his financial partner, his
cheerleader, his business partner, his housekeeper, and at times his
tormenter. My feelings for him run deep -1 love him like he is a part
of my soul. [He] is my cheerleader, my positive thinker, my
mechanical mind while I am his hands, my partner, my scapegoat for
my sad or anxious feelings, (wife of man with spinal cord injury)
Introduction
SCI typically causes immediate and usually permanent manifestations: an alteration
or loss of muscle function and sensation in the body below the level of the injury, as
well as impairment in bowel and bladder emptying and sexual function. These
conditions alone mean that the injured individual often requires assistance from others
to carry out day-to-day functions such as using the toilet, dressing, bathing, or getting
out of bed.
Other body systems also are likely to be affected. For example, the respiratory system
in people with injuries of the upper thoracic and cervical spine is often compromised
due to the lack of adequate musculature to produce a strong cough or inhale deeply.
People with high cervical injuries in whom the diaphragm is partially or wholly
paralyzed often require full- or part-time ventilator support. In addition, the absence
6


or alteration of sensation may produce problems with skin breakdown due to pressure
over a bony prominence or skin trauma from performing certain mobility activities
such as transferring from the wheelchair to the car or toilet. Osteoporosis due to
paralysis and disuse is commonly felt to be an underlying risk factor for pathologic
fractures following spinal cord injury. Lower extremity osteoporosis develops rapidly
in the first post-injury year, with a loss of about one-third of the original bone mass by
16 months post-injury, increasing the likelihood of fractures (Garland et al., 1992).
Just as spinal cord injury can have a major effect on organ system functioning in the
acute injury phase, evolving post-injury physiology leaves SCI survivors with
autonomic and somatic nervous system dysfunction, which results in lasting
impairment of many organ systems. The cumulative effect of this pathophysiology
over many years of survival may result in the development of secondary
complications, which will become increasingly prevalent with a longer duration of
impairment (Charlifue & Lammertse, 2002). These conditions may, in turn, have an
impact on the ability to remain independent in a variety of activities, and assistance
from others becomes critical. Because the raison detre of clinicians and
professionals involved in SCI rehabilitation is to reduce disability by increasing the
independence with which individuals perform activities (Ditunno, Cohen, Formal, &
Whiteneck, 1995), enabling them to live in the community (Dijkers, Buda Abela,
7




Gans, & Gordon, 1995), individuals optimally will receive needed caregiving in the
home. In fact, over 91% of all people with SCI are discharged to home settings after
initial inpatient acute care and rehabilitation (Eastwood, Hagglund, Ragnarsson,
Gordon, & Marino, 1999).1 Evidence suggests that in the years following
rehabilitation, people with SCI will have an increasing need for help as they age
(Gerhart, Bergstrom, Charlifue, Menter, & Whiteneck, 1993). This, coupled with the
fact that 40% of all people with SCI are now over the age of 45 and one-fourth have
been injured for 20 or more years (Menter, 1993), makes it clear that the number of
people who require assistance will only increase in the coming years.
The Nature of Caregiving
In the United States, most home care is provided through informal channels by family
members (Mintz, 1996). Similarly, almost 70% of people with SCI the majority of
them men receive some form of assistance and support from family members
(Nosek, 1990). In 1985, Schulz and Decker reported that 41% of people with SCI
named only one support person, most frequently a female spouse. One individual,
therefore, very frequently assumes most or all of the responsibilities of caring for a
Depending on factors such as the neurological severity of injury, funding, age, equipment provision or
concomitant complications, the length of inpatient rehabilitation hospitalization can vary; however, the
2002 reported mean length of stay within the SCI Model Systems ranged between 7-36 days for acute
care and 18-73 days for rehabilitation (National Spinal Cord Injury Statistical Center, 2003).
8


person with a disability (Decker, Schultz, & Wood, 1989). This responsibility carries
with it several physical and emotional risks.
Pearlin and colleagues (1990) have identified a caregiving model that measures
constructs such as overload (the caregivers sense of being overwhelmed by tasks);
loss of self (the degree to which the caregiver feels a personal loss of his/her own
identity during caregiving); management of meaning (the degree to which respondents
are able to find meaning in their situations); and expressive support (the degree of
help and support from friends and relatives. They describe caregiving as the
behavioral expression of ones commitment to the welfare of another, whereas caring
is the affective component. Under this definition, all caring relationships include
some degree of assisting one another. However, as the authors indicate, when serious
or prolonged impairment figures into the relationship, caregiving may become the
dominant component of the interactions and may, under some circumstances, expand
to the point where it occupies virtually the entirety of the relationship (Pearlin et al.,
1990, p. 583). Consistent with the dominance of caregiving over other life
experiences, at least some of those who provide assistance to loved ones report
diminished social activities, less personal time, lack of privacy, and increased anger
and depression (Given, Collins, & Given, 1988).
9


The most commonly studied themes of caregiving focus on stress, depression, and
burden, which are discussed in detail below. The vast majority of research regarding
spouse caregiving has been done in the area of Alzheimers disease, with a primary
focus on caregiver stress. However, as interest and awareness of caregiver needs
increase, investigators have started looking at the multifactorial nature of caregiver
distress and caregiver life satisfaction. The following review of the literature
discusses caregiving and disability in this broader context, particularly examining the
emotional and social facets of caregiving. Caregiving research in spinal cord injury is
described separately, as many of the issues of caregiving (and hence, the ways in
which they can be studied) likely differ. Specifically, this review examines several
studies and describes the basis of each on underlying theories of stress, depression, or
burden. Tables 1 and 2 review the breadth of findings in the literature.
10


Table 1
Summary of Relevant Literature Non-SGI Studies
udy Information Study Sample Findings Limitations
RDEN ;
('nor 1990 e Control Design 125 female spouses of men with neurological diagnoses; control group of 30 with healthy spouses Extended time in caregiving contributed to more perceived burden, more illness episodes Reports of stress based on comments from participants; no attempt to tie qualitative and quantitative findings
-nsch & Spa id 9 week training gram for elderly ase caregivers; pre- vention data 131 caregivers (70% female; 89% white); mean age 69 years; average length of marriage 39 years. 51% of care receivers had dementia; others included multiple diagnoses limiting activities of daily living Men and women reported different types of formal services used; men more likely to receive home help. Women more likely to participate in outside activities, use counseling and other services; few gender differences in use of informal support. Women reported more coping problems and more subjective burden than men. Memory and behavior problems key predictors Small proportion of males, mixed diagnoses mal generalizations difficult; coping measured in absence of any stress assessment; possible bias < to recruitment strategy (volunteers identified through self-referral or agency-referral)
it et al 1986 ) year follow-up n earlier study; in- te interviews 64 spouses (33 female; 31 male) of people with Alzheimers disease; mean ages 72.27 male and 63.37 female; generally well-educated, financially comfortable Compared burden, social support, quality of relationship in couples with partner living at home vs living in institution. Burden lower for wives at follow-up compared to initially; also lower for women who had placed spouse in nursing home. Social support not a factor in placement decision Factors other than caregiver burden may have le to placement decision; multiple t-tests in analysi may be limitation, small sample size; variety of factors may have contributed to change in responses between time 1 and 2
*ESS t r ~ f * I * :
ler et al 1995 a from a study of iience of race and der on spouse igivers 215 spouses of people with Alzheimers disease (64% female, 64% white); mean age 74.7 years Role strain (stress) and depression highly correlated. Stressor variables included spouses limitations in physical ADLs, caregiver reactions to disruptive behavior, and caregiver self-reported health. Depression predicted by the above plus lower sense of control and lower appraisal of caregiving confidence. Comparatively small proportion of African American males; difficult to interpret cultural differences; sample is quite elderly and findings may not be generalizable to younger caregivers
ins et al 1992 iy to assess itive, neutral and ative aspects of igiving 181 primary caregivers of recently hospitalized veterans who required 2+ hours of assistance. Mean caregiver age 50.5; mean months of care provision 37.5. Caregivers exhibited more psychological symptoms on brief symptom inventory compared to population norms; reported family functioning levels similar to normal population; caregiver adjustment related to family communication and problem solving, caregiver subjective health and length of time providing care No theoretical framework on which to base wort not truly a study of stress per se, but an assessmi of stressors, with family functioning as outcome Biased toward experiences of female caregivers


Table 1 (continued)
Summary of Relevant Literature Non-SCI Studies
idy Information Study Sample Findings Limitations
ESS(cont)
is Stephens ct al ss stressors and rds of caregiving 95 daughters, daughters-in-law of parents needing help with at least one of three activities of daily living; caregivers also occupying roles of mother and wife. Mean age 43.9 years, 94% white Rank ordering of stressors and rewards for each role. Stress in at least one role accounted for significant portions of variance in well-being scores. Role stress scores related to poorer well-being and reward scores related to better well-being. Higher stress in caregiver role alone related to better well-being than for those with stress in all three roles. Small sample size, selection bias (volunteers), limiter to daughters, so results could not be generalized to caregiving spouses. Need to examine role quality, nc just role occupancy
XTIinnd RESSION. . & f %# gr~, ,- { '( M -* 'A \ 1
lane et al 1997 > caregivers ;itudinal study 642 informal AIDS caregivers; mean age 40 years; 368 with longitudinal follow-up, 28% caregivers seropositive at time 1, 67% positive at time 2 Reports of poor health increased as assistance with activities increased. Care-related stress related to physical health of caregiver; care recipient problem behaviors, caregiver work strain and unemployment associated with poorer caregiver health. Role overload significant + correlate of poor health in seropositive caregivers. Seronegative males report better health than seronegative females. Caregiver depression a correlate of poor caregiver health. Due to unique nature of AIDS, high proportion of seropositive caregivers, and relatively young age of caregivers, results difficult to generalize to other samples of caregivers
;ht et al 1998 cipants in survey omens health s compared with carers; attempt to rate stress, ession, burden 742 women in midlife, of whom 12% involved in care of another person with chronic health disorder. Mean age of the 90 caregivers was 48.44, looking after a total of 121 people No significant elevation in rating of depression. Caring under some circumstances (e.g. single parent not living with caregiver) more stressful than under others (both parents not living with caregiver). Significant predictors of high burden scores were poor subjective health ratings, co-residence with the carer, lower satisfaction with social support. Small sample size of caregivers; variety of relationships to care receivers with very small numbe of spouse carers. Mailed survey but no mention of response bias (742/980 received)
ieet al 1998 ;s process model edict physical and :ai health itudinal 122 caregivers of dementia patients; 70% female, 60% white, 49% spouses, 30% daughters, 21% other relatives. Mean caregiver age 59.57 years Changes in appraisals of stressfulness positively relate to changes in depression and physical health symptoms. Changes in depression positively correlated with changes in health symptoms. Patient deterioration over time did not directly predict changes in mental or physical health outcomes. Participant attrition over time, unable to assess outcomes in caregivers who had lost or had to place care recipient; mixture of caregiver relationships. Study of dementia patients only cant generalize to physical impairments


Table 2
Summary of Relevant Literature SCI Studies
tudy Information Study Sample Findings Limitations
cker ct al 1989 dy to identify erminants of life sfaction and iression in primary egivers of middle- d and elderly iple with SCI 67 primary caregivers; 88% of caregivers female, mean age 53.6 years; 85% spouses of person with SCI Caregivers and people with SCI scored similarly on measure of life satisfaction but slightly lower than general adult public. No significant differences in depression scores between caregivers and general population. Caregivers with better perception of own health are less depressed and have better life satisfaction. Greater levels of social support, greater sense of control over their lives, and less perceived burden associated with higher levels of life satisfaction and less depression. Small samples size; no assessment of stress. Burde assessed by single-item question, not by validated instrument. Mixture of spouses and other family caregivers.
ifzcnkamp et al >7 tish longitudinal dy of aging with 1 124 spouses of people with long- term SCI; 90% female, mean age 52 years; mean length of marriage 29 years, 52% married pre-injury, 64% perform some caregiving activities Spouses scored significantly higher than partner with SCI on measures of depression. Spouses scored higher on measures of stress. No difference in well-being scores. Significantly higher proportion of caregiving spouses reported physical stress, emotional stress, burnout, fatigue, anger and resentment compared to noncaregiving spouses. Actual scores on measure of perceived stress did not differ. Caregiving spouses had significantly higher depression scores. No assessment of marital strength or marital adjustment; mixture of pre- and post-injury marriages; small sample size; British sample makes generalization to US caregiving samples difficult
iwchuk et al 1998 idy of caregiver comes in first year it-SCl 62 caregivers for people with recent SCI; 87% female, mean age 40.1 years, assessed during hospitalization, at 1,6 and 12 months post-discharge Physical symptoms of caregivers positively related to initial status of depressive behavior. Initial level of positive affect inversely related to depressive behavior. Rate of change in depressive behavior associated with expressive support. Anxiety was best predictor of variability in initial levels of physical symptoms. Physical symptoms statistically significant in predicting variability in anxiety and depressive behavior. Younger patient age associated with more anxiety at initial measurement and low positive affect associated with higher initial depression scores. No mention of relationship between person with SC and caregiver (spouse? parent?); small sample size; caregiving assessed early in injury course no indication of long-term changes/characteristics; stre not measured objectively as a construct, inferred frr measure of anxiety.


Caregiver Studies with Non-SCI Samples
Disability, whether its onset is slow and progressive or sudden, affects not only the
individual but the family as well, particularly when they are called upon to provide
help to their loved one. The degree of impact, however, may vary. The literature
does not provide a clear conceptualization of what constitutes caregiver distress. In
several studies, investigators have focused on caregiving stress and depression, and
their association with a number of factors such as the caregivers sense of control
(Miller et al., 1995), family functioning (Evans et al., 1992), and the severity of
disability (Zarit et al., 1986).
Another component of the caregiver role is the evaluation of burden, or role strain.
While depression, burden and stress may be closely related, researchers often treat
them as separate constructs and assess them with different measurement tools. The
following review looks at these various types of caregiver distress.
Burden
Zarit and colleagues (1986) defined burden as the extent to which caregivers perceive
their emotional or physical health, social life, and financial status as suffering as a
result of caring for their relatives. Burden includes many factors outside the
caregiving role, such as family commitments, employment, and other demands that
14


compete for the caregivers time and energy. Bush and Job (1993) reported that
nearly one fourth of the caregivers in the US work the equivalent of one and one half
jobs when employment and caregiving responsibilities are combined. Even in the
absence of paid employment, informal caregivers take on additional tasks not
normally associated with their familial or spousal roles. Caregivers often must
provide home health care, assume responsibility for most or all of the household
chores, and become the primary wage earner, the primary or only parent, and the
healthcare gatekeeper (Holicky, 1996). Caregivers may find that these increased
responsibilities leave little time for their own needs. According to Bush and Job
(1993), compensating for the emotional drain of constant responsibility is one
example of invisible (unobservable) work. This type of activity is difficult to
quantify, but its presence can undoubtedly create added burdens on the caregiver.
Other examples of such unobservable work include resolving guilt about negative
feelings toward the person to whom assistance is provided, coping with upsetting
behavior of the dependent person, working through role reversal in the lifelong
relationship between caregiver and care recipient and satisfying the need for creativity
and originality to offset tedious routines. Gaynor (1990), in a study of 155
caregivers, reported that women who had been involved in caregiving for an extended
period of time had greater feelings of perceived burden, and the length of time one
15


had engaged in home caregiving was associated with an increase in illness episodes
and/or perceived burden in the caregiver.
Not only is burden theoretically viewed as a factor in caregiver strain, it is measured
with several instruments. Barusch and Spaid (1989) studied 131 caregivers (92
women) providing assistance to family members with various diagnoses, including
dementia, stroke, lung disease and general frailty. Caregivers responded to items that
measured the number of specific tasks the caregiver performed for the care recipient
and the method of coping with caregiver problems. An inventory that presented 34
situations that might pose problems for the caregiver, scored using a five point Likert
scale of coping effectiveness, assessed caregiver problems and coping. Caregiver
burden was measured using the Zarit et al 20-item burden scale (Zarit, Reever, &
Bach-Peterson, 1980). Women reported more coping problems than did men, but few
differences in coping styles were identified. In addition, women caregivers reported
experiencing significantly higher subjective burden than did men (p <.001), with the
most important predictive variable being the number of memory or behavior problems
exhibited by the care recipient. The second most important factor, caregiver age,
decreased burden. That is, younger caregivers reported greater perceived burden.
16


Zarit and colleagues (1986) reported the amount of burden caregivers felt depended
on the degree to which they coped well or poorly with specific problems related to the
care recipients disabilities and the resources they had available, especially for
obtaining assistance and occasional relief from caregiving tasks. They also pointed
out that while there has been little theory to guide research, it is frequently assumed
that the caregivers burden is strongly related to the severity of the patients
disabilities (p. 260). In this two-year longitudinal investigation of 64 caregiving
spouses, they found that subjective factors (such as perceived emotional or physical
health, social life, and perceived suffering as a result of providing care) on the part of
the caregiver were more important than the severity of the care recipients illness in
determining burden. The authors concluded that their findings regarding burden were
consistent with general buffering models of stress, which propose that the impact of
harmful events is mediated by whether subjects perceive them as harmful and by their
own coping responses (Lazarus & Folkman, 1984b).
The variety of tools used to assess burden often differ substantially in their
conceptualization of burden, with some instruments providing a simple total score
that is intended to reflect burden (Zarit, Reever, et al., 1980), and others using a
multidimensional approach incorporating component subscales to assess different
aspects of burden (Novak & Guest, 1989). Furthermore, the majority of validated
17


burden assessments have been developed for and tested with samples of people
providing assistance to individuals with cognitive disabilities. In order to evaluate
burden as an outcome in other disability groups, therefore, the multidimensional
measures may show more promise, as the cognitive component scores can be
interpreted separately or, if appropriate, disregarded.
Clearly, studies of caregiving highlight the variety of factors that may contribute to
perceived burden, including physical status of the care recipient, duration of
disability, multiple role responsibilities, and availability of social support systems. In
spite of potential buffers and rewards of caregiving, however, the burden literature is
replete with findings associating perceived burden with increased stress, depression,
and other negative outcomes. It is critical, then, to consider the interrelationship of
these other outcomes and their possible predictors in caregiver research.
Stress
Stress is a condition that results when an individual has a perceived discrepancy
between the demands of a situation and the resources of that persons biological,
psychological, or social systems (Sarafino, 1990).
18


Caregiver stress refers to ongoing problems within the caregiving role that cause
disruptions or changes that have potential to tax or exceed the caregivers resources
(O'Brien, 1993). Pearlin and colleagues have conceptualized a stress process model
of Alzheimers caregivers, illustrated in Figure 1.
Figure 1
Alzheimers Caregiver Stress Model
This model comprises four domains, the first of which is the background and context
of distress, which includes factors such as socioeconomic characteristics, caregiving
history, and composition of the family and network. The second domain, the
stressors, includes primary stressors, which encompass the demands of caregiving,
(problematic behavior, dependencies in daily activities) and the perceived hardships
19


experienced by the caregiver that can manifest as fatigue and overload. Also included
in the second domain are the secondary stressors of role strains (family conflicts,
work/caregiving conflicts, economic problems, and social life limitations), and
intrapsychic strains (self-esteem, mastery, role captivity and competence). The third
domain involves mediators of stress, such as coping skills and social support. Finally,
the fourth domain represents the outcomes of stress, which include depression,
anxiety and difficulties with physical health. While conceptually useful, this model
only suggests a one-way influence of mediators on the stressors it does not address
the impact of the various stressors on possibly breaking down the coping capabilities
or social support network of the caregiver.
In addition to the stress process model, stress and coping models provide a focus for
much of the caregiving stress research. In these models, combinations of stressors
(e.g., perceived burden) and resources (e.g., available social support or finances)
influence an individuals response to stress (Lazarus & Folkman, 1984a). This model
is illustrated in Figure 2. In this example, a stressor is evaluated in terms of its
significance or threat (primary appraisal). If the threat is considered minimal or if the
consequences of the stressor are potentially positive, the individual would not
perceive this stressor as a threat. In the process of secondary appraisal, individuals
evaluate their ability to control or change the situation, manage their own reactions
20


and evaluate their coping resources. If the resources or sense of control are adequate,
the threat is reduced. These appraisal processes, then, contribute to ones perception
of the stressor. The entire process is fluid, and consists of frequent re-appraisal of the
stressors and resources.
Miller and colleagues (1995) used the stress and coping theoretical framework when
examining issues of control as they relate to caregiver distress. In their study of 215
spouses of people with Alzheimers disease, they looked at predictors of caregiver
depression or role strain. Stressor variables included the spouses limitations in
physical activities of daily living (PADLs) such as dressing or grooming, task distress
(the amount of effort or difficulty experienced by caregivers in the performance of
caregiving-related tasks), behavior problem upset (caregiver reactions to disruptive
behavior, such as emotional lability or aggressive behavior), and the caregivers self-
report of physical health. To assess psychological resources, the investigators
examined the caregivers sense of control, based on the Mirowsky and Ross
conceptualization (Mirowsky & Ross, 1990). This measure compares claims of
responsibility for good or bad outcomes (instrumental attributions) with claims
denying responsibility for good or bad outcomes (fatalistic attributions), and assesses
ones sense of personal responsibility for ones own life and sense of control over
others.
21


Figure 2
Stress Appraisal Model (based on Lazarus & Folkman 1984a)
PRIMARY SECONDARY
APPRAISAL APPRAISAL
^consequences * resources
*harm *ability to cope
1
Positive Consequences
Little/No Harm
1
Not a threat
1
Adequate Resources
Good Ability to Cope
1
Threat reduced
Finally, in this same study, caregiver mastery ones general feeling of competence in
caregiving was examined (Miller et al., 1995). Significant predictors of depression
included poorer perceived health, greater behavior problem upset, higher levels of
PADL impairments in the spouse and greater task distress, a lower sense of control
and lower appraisal of competence in caregiving. They also found high positive
correlations between role strain and depression scores. In addition, they reported that
a sense of control and caregiver mastery moderated the effects of stressors on
depression. A highlight of this particular investigation was the multifactorial
approach to caregiver distress; however, they did not attempt to assess the impact of
22


other variables that could shed light on factors mitigating caregiver stress, such as
quality of the caregiving relationship.
Evans and colleagues (1992), in their investigation of caregiver stress, did address
issues of relationships, but did not identify a particular theoretical framework on
which to base their work. Rather, they illustrated some of the gaps in existing
research most notably, the focus on the negative aspects of caregiving. In their
work, they evaluated 181 caregivers for psychological adjustment and family
functioning. Their findings supported the notion that family functioning, described
partly as communication and problem solving, reliably predicted caregiver
adjustment.
Gender also appears to play a role in the perceived degree of caregiver distress, with a
predominant theme of greater depression among caregiving wives than caregiving
husbands (Alspaugh, Stephens, Townsend, Zarit, & Greene, 1999; Beeson, Horton-
Deutsch, Farran, & Neundorfer, 2000; Neundorfer et al., 2001; Beeson, 2003). In
addition, wives who provide assistance to their husbands with a disability report
greater depression than non-caregiving spouses (Beeson, 2003). A population-based
study of 1002 male and female informal caregivers found that women were more
likely than men to experience greater emotional challenges associated with caregiving
23


and attempting to balance other family and work obligations (Navaie-Waliser,
Spriggs, & Feldman, 2002). Women in this study were more likely than men
caregivers to deliver more intensive care such as providing assistance with bathing,
dressing, and managing incontinence.
In an effort to examine both the negative and potentially positive aspects of
caregiving, Parris Stephens and colleagues (1994) examined role stress and rewards
as predictors of well-being. They studied 95 women simultaneously occupying the
roles of caregiver, mother and wife, looking at both the competing-demands
hypothesis (Brody, 1981) and the expansion hypothesis (Marks, 1977). The former
hypothesis assumes that the roles people (specifically women) occupy in addition to
that of caregiver compete for their time and energy, and such responsibilities make
them more vulnerable to stress. Conversely, the latter posits that energy gains, rather
than energy expenditures, are accrued by individuals with multiple roles, predicting
positive consequences of multiple roles. The investigators measured stress and
reward in three arenas of the womens lives caregiver, mother, and wife. A single
self-rating item measured the physical health of the respondents. Role overload was
also assessed with a single item, rated on a five point Likert-type scale. Stress in at
least one role accounted for significant portions of variance in well-being scores.
Role rewards explained additional variance and were significant predictors of well-
24


being even when controlling for role stress. In addition, women experiencing higher
levels of stress in their caregiver role alone demonstrated better well-being than did
the women experiencing larger amounts of stress in all three roles. It is unfortunate
that this studys limitations included small sample size, selection bias of the sample,
and a cross-sectional design that precluded interpretations of any causal links between
role stress, role rewards and well-being. It did, however, provide a valuable
contribution to our understanding of the impact of multiple roles on caregiver well-
being, and highlighted positive factors related to the caregiving role in its relationship
to other life roles.
The above studies, like those in the burden literature, demonstrate the
multidimensional nature of caregiver stress. The concept of buffers or mediating
factors is more clearly illustrated using the stress process models. However, the
measurement of stress as a unique construct is not consistent, and is often an assumed
outcome based on more objective measures of burden, depression, physical health of
the caregiver, and tasks performed. Rarely is stress, itself, measured. Rather, it is
interpreted from measures of anxiety, burden, depression, or other manifestations of
distress. In order to evaluate stress as a specific outcome and to associate it with other
outcomes and predictors, a measurement tool designed to address stress specifically
appears needed.
25


Caregiver Health and Depression
A particular concern for caregivers is the risk of injury incurred while performing
caregiving tasks. Numerous studies have identified caregiving as a risk factor for
health problems, including more episodes of infectious illnesses (Kiecolt-Glaser,
Dura, Speicher, Trask, & Glaser, 1991), poorer immune responses to vaccines for flu
and pneumonia (Kiecolt-Glaser, Glaser, Gravenstein, Malarkey, & Sheridan, 1996;
Glaser, Sheridan, Malarkey, MacCallum, & Kiecolt-Glaser, 2000; Vedhara et al.,
1999), delayed or slowed wound healing (Kiecolt-Glaser, Marucha, Malarkey,
Mercado, & Glaser, 1995), development of mild hypertension (Shaw et ah, 1999)
(Grant et ah, 2002), coronary heart disease (Vitaliano et ah, 2002) and overproduction
of the proinflammatory cytokine, IL-6 (Kiecolt-Glaser et ah, 2003). Schultz et ah
(1990) reported that it could reasonably be predicted that caregivers should have
higher rates of physical illness, since they are chronically exposed to high levels of
stress. Even in the absence of actual bodily injury or illness, caregiving may
negatively affect perceived health, often manifested as role strain, stress or
depression. Symptoms and health problems related to caregiving include
hypertension, arthritis, back pain, stomach upset or indigestion, and cardiac problems
(Bush & Job, 1993) and a generalized decline in perceived health (Douglas & Daly,
2003). Furthermore, Gaynor (1990) reported that caregivers tend to ignore their own
26


health and are more apt to forego health promoting behaviors if those behaviors
conflict with caregiving, and thus end up in a cycle of neglect.
LeBlanc and colleagues (1997) investigated the effect of stress on health among
caregivers of persons with AIDS using the stress proliferation framework of Pearlin
and associates (Pearlin, 1989; Pearlin et al., 1990). In this model, primary stressors
are defined as difficulties associated with the actual tasks of caregiving, and stem
directly from the needs of the care recipient and the nature and magnitude of the care
demanded by these needs. Secondary stressors, on the other hand, are problems that
arise as a consequence of providing care, but do not directly entail the tasks of
caregiving. These may include factors such as family and job conflicts, financial
problems, or loss of self-esteem. The secondary stressors are generated and
exacerbated by the primary demands of caregiving in a process of stress proliferation.
The LeBlanc study asked a sample of 642 informal caregivers of persons with AIDS
to self-assess their frequency of experiencing symptoms of eight common ailments
(headaches, cold/chills/sore throat, indigestion, constipation, diarrhea, weakness,
backaches, shortness of breath). Six items from an established symptoms checklist
assessed depression. In addition, an array of care-related stressors were measured,
including perceived stressors arising from the provision of help, stressors related to
personal financial strain and conflicting demands between work and caregiving. As
27


assistance with activities increased, so did reports of poor health. Caregiver
depression was also a significant correlate of poor physical health.
In contrast, Knight and colleagues (1998) surveyed 742 women, of whom 90 were
engaged in caregiving for a family member with a traumatic brain injury. They found
no significant elevation in subjective ratings of depression comparing the caregivers
to those who did not report having a caregiving role, suggesting that caregiving is not
necessarily deleterious to health and well-being.
Goode and associates (1998) also used a stress process model to describe the role of
primary stressors on caregiver health, positing that a number of processes can occur
over time to explain changes in caregiver mental and physical health outcomes. To
assess the interplay of numerous factors, they examined changes in primary
caregiving stressors, appraisal, coping, social support and mental and physical health
outcomes over a one year period. Their 122 participants reported caregiving stressors
as they related to the care recipients impairments and need for assistance. As the
authors expected, changes in appraisal of stressfulness were positively related to
changes in depression and physical health symptoms, and changes in depression were
positively correlated with changes in health symptoms. These findings supported the
usefulness of a stress process model, demonstrating that psychosocial resource
28


variables (appraisal, coping responses, and social support) predicted longitudinal
changes in caregiver mental and physical health.
A significant consequence of caregivers neglect of their own physical and emotional
health is their increased risk of becoming injured or too ill to provide assistance to
their dependent partners. Their own perceived poor health, in turn, may lead them to
report more depression and less life satisfaction (Decker et al., 1989), and render them
less able to cope with their various responsibilities. When caregivers are unable to
cope effectively with role responsibilities, the health and well-being of the care
recipient also may be jeopardized (Elliott, Shewchuk, & Richards, 1999). Ultimately,
the inability to continue providing care because of declining physical or emotional
health may lead to institutionalization of the care recipient (Burton, Newsom, Schulz,
Hirsch, & German, 1997). Institutionalization may be considered an acceptable
option for frail elderly individuals who have cognitive impairments, such as with
Alzheimers disease. Individuals with spina] cord injury, who may be younger and
still have many years of life, however, may find it less desirable, both for themselves
and for their caregiving spouses.
The literature regarding physical injury, illness and depression is almost inextricably
interwoven with reports of caregiver stress and burden. However, in the non-SCI
29


caregiver literature, depression is rarely measured separately. Again, in order to fully
evaluate depression and its unique relationship to other caregiver outcomes, as well as
the factors which may be predictive of caregiver depression, it would need to be
measured with a well-validated tool. A significant contribution to the caregiver
literature, therefore, would be to treat these distinct outcomes both independently and
in relation to one another, and use well-validated instruments designed to evaluate the
varying aspects of caregiver burden, stress and depression.
Caregiver Studies with SCI Samples
Many findings from the extensive caregiving literature cannot easily be transferred to
caregivers of people with SCI for three reasons:
1) Most studies focus on caregivers of people with Alzheimers disease or dementia.
Because these caregiver stress/depression findings typically are associated with
the level of cognitive function of the care recipient (Miller & Cafasso, 1992;
Marchi-Jones, Murphy, & Rousseau, 1996), they may not apply to individuals
with SCI, whose care needs are more typically associated with physical limitations
than with cognitive deficits. Because caring for a person with SCI is likely to
entail a great deal of lifting during transfers to and from bed, wheelchair and car,
SCI caregivers may be at risk for incurring injury. Therefore, those studies that
30


report an association between the physical act of caring and emotional stress or
depression, may possibly permit generalizations to SCI caregivers.
2) The mean caregiver age in the majority of studies is 60 years or older. The
highest occurrence of SCI is in the 16-30 age group (National Spinal Cord Injury
Statistical Center, 2003). In addition, younger caregivers of partners with
dementia are more unhappy and resentful of the caregiving role than are older
individuals (Fitting, Rabins, Lucas, & Eastham, 1986). Younger spouses
providing care to their younger partners with SCI may therefore demonstrate
higher levels of depression or stress than reported in the overall caregiver
literature.
3) Alzheimers disease and other pathologies that lead to a need for care typically
develop gradually over time, possibly allowing spouse caregivers an opportunity
to adjust to changes in function and behaviors of their partners. Furthermore,
many of these disease processes generally result in death within one decade of
onset (Guttman, Altman, & Nielsen, 1999; Wolfson et al., 2001; Alzheimer's
Association, 2003; eMedicine Consumer Health, 2003). Conversely, traumatic
spinal cord injury results in an immediate change in functional status and need for
assistance. Furthermore, the life expectancy for individuals with SCI has been
steadily increasing over the last five decades (Frankel et al., 1998), and many
people can expect to live several decades with SCI. Caregivers, therefore, may be
31


in a position of providing assistance for lengthy periods of time and more likely to
experience stress, depression or incur physical injury from performing caregiving
tasks over several years.
Because SCI survivors and their spouses are younger at the onset of disability and
have different caregiving needs than older individuals with disabilities, the existing
models of caregiving may not apply (Weitzenkamp et al., 1997). Therefore, studying
caregiving in SCI provides us with an excellent opportunity to explore the long-term
consequences of caregiving. It is unfortunate that there is relatively little research
examining caregiving in SCI; and most SCI studies are limited by small sample sizes
or combine caregiving spouses with other caregiving family members. However, in
spite of such limitations, several interesting findings are reported.
In a study of 67 primary caregivers (57 spouses and 10 other relatives) of middle aged
and elderly individuals with SCI, Decker and colleagues (1989) found no statistically
significant differences in levels of depression between caregivers and the general
population. They examined numerous theoretical constructs, including perceived
health status, perceived control and perceived burden. Caregivers were also asked to
assess the burden of caring for the person with disability. This study found that
individuals with a perception of better overall general health status had less
32


depression and greater life satisfaction. Greater perceived control over their lives was
also associated with higher levels of life satisfaction and less depression. Those who
perceived greater burden were more likely to have more depression and diminished
life satisfaction.
According to Stull et al. (1994), the absence of a comparison group is a problem with
caregiver burden research in all of the studies described previously. Family members
who are not caregivers do not experience the caregiving burden, which makes
comparisons between caregivers and noncaregivers impossible. Studies of burden
require caregivers to assess the amount of time spent on caregiving and rate their level
of burden accordingly; based on the assumption that if they spend many hours each
day on caregiving, they must be burdened. The authors, therefore, suggest that using
measures of well-being instead of burden makes it possible to compare caregivers
with noncaregivers.
A recent study of spouses whose partners have lived many years with SCI attempted
to address this gap. Weitzenkamp and colleagues (1997) investigated 124 British
married couples in which one partner had a spinal cord injury of at least 20 years
duration. Comparisons of caregiving spouses with non-caregiving spouses found
overall depression scores as measured by the Center for Epidemiological Studies
33


Depression Scale (CES-D) significantly higher for the former group than the latter.
Stress scores, however, were not significantly different, although caregiving spouses
reported greater stress than those not providing care. Although there was a strong
overlap between the SCI survivors level (severity) of injury, and their resulting
dependency, this factor failed to explain the differences in stress scores between
spouses who provided care and those who did not. In addition, the study found that
significantly more caregiving spouses reported feelings of burnout, fatigue, anger and
resentment than did their noncaregiving counterparts. The caregiving spouses also
scored significantly higher on measures of depressive affect, and reported higher
scores on measures of stress compared to noncaregivers (Weitzenkamp et al., 1997).
Since this study sample was British, and no similar studies with U.S. samples have
been conducted, it is not possible to generalize these findings to U.S. SCI caregiver
samples.
Shewchuk and colleagues (1998) examined the dynamic processes involved in health
and adjustment of family caregivers of people with SCI in the early time period post-
injury. They assessed 62 caregivers (8 men, 54 women) of SCI survivors during
initial rehabilitation and at one, six and 12 months post-discharge. Caregiving
experiences were measured using Pearlins model of caregiving (Pearlin et al., 1990).
Depressive behavior was measured with the CES-D. Physical symptoms were
34


assessed, as were anxiety and positive/negative affect. Lower levels of expressive
support the degree of help and support from friends and relatives contributed to
the development of depressive behavior and anxiety in caregivers over time.
Overload, loss of self, and management of meaning were unrelated to the
development of distress. Based on changes in depression scores on the CES-D
between the initial and final assessments, this study found an association between
increased depression and both younger patient age and low positive affect of the
caregivers. All of these findings suggest increasing depressive symptomatology as
time passes, which is counter to clinical expectations, according to the authors.
Unfortunately, no information is available in this or any of the caregiver studies
described above regarding the amount of time spent caregiving and the impact that
extra caregiving has on stress, depression, or burden. Although Weitzenkamp and
colleagues (1997) dichotomized their sample into caregiving and non-caregiving
spouses, they did not explore differences in outcomes based on the extent of care
provided.
Studies of caregiving in general, and specifically in SCI, have examined caregiver
distress and strain within the context of various stress or burden models. Few other
theoretical avenues have been explored, particularly those that evaluate the nature of
the caregiver-care recipient relationship as a component of caregiver distress. Clearly,
35


the quality of such relationships may be a meaningful factor in explaining caregiver
distress, making exploration of alternative theories a necessary and compelling
option.
Attachment Theory
Because of the interplay of numerous factors affecting caregiver health and distress,
designing theoretically-driven research is extremely complex, and it is unlikely that a
single theory or model can adequately address all salient issues. However, one theory
that may provide the glue to understanding the multidimensional issues related to
caregiving both the positive and the negative is attachment theory.
Adult attachment theories developed from the idea that attachment is a universal,
enduring affective tie between a given child and a specified adult (Bowlby, 1980).
More broadly, Bowlby suggested that attachment theory is a way of conceptualizing
the propensity of human beings to make strong affectional bonds to particular others
and of explaining the many forms of emotional distress and personality disturbance
including anxiety, anger, depression and emotional detachment, to which unwilling
separation and loss give rise (Bowlby, 1977). Expanding this concept of attachment
throughout ones life, Bowlby posited that there is a strong causal relationship
between an individuals experiences with his parents and his later capacity to make
36


affectional bonds. By affectional bonds, Bowlby was referring to particularly close
ties in which the partner is important as a unique individual and interchangeable with
none other (Ainsworth, 1989). Specifically, Bowlby suggested that the behavior of
parents, and of anyone else in a caregiving role (emphasis added), is complementary
to attachment behavior (Bowlby, 1977).
Insight into how early attachment and subsequent romantic relationships are linked
grew out of the work of Mary Ainsworth, a colleague of Bowlbys. She stated that
attachments occur at all ages, and these attachments are durable, even under the
impact of adverse conditions (Ainsworth, 1969). Influenced by Bowlby, Ainsworth
expanded the theory in her studies of infant-mother relationships in Uganda and
Baltimore. It was from this work that she identified three attachment types secure,
insecure (anxious), and not-yet attached (ambivalent) (Waters & Cummings, 2000).
Useful to the examination of spousal caregiving relationships is Ainsworths assertion
that attachment is a synonym of love, dependency is not and posed the question,
Does a phobic wife who clings to her husband and constantly
seeks his proximity love him more than a woman who is less
neurotic and more competent loves her husband? The answer is
clearly, No, not necessarily. Indeed, a clinician might infer
that the phobic wife loves less than the healthier wife.
(Ainsworth, 1969, p. 39).
37


While much of the work discussed by Bowlby and Ainsworth focuses on infant
attachment, others have suggested that attachment theory has relevance across the life
span (Waters & Cummings, 2000). Hazan and Shaver (1987) carried out further work
regarding adult attachment. They theorized that adult romantic relationships differ
from infant-caregiver relationships in some significant ways; romantic love, however,
is fundamentally an attachment process through which affectional bonds are formed
(Simpson & Rholes, 1997). More work on adult attachment has grown out of a body
of research based on Hazan and Shavers proposed tripartite classification of adult
attachment styles in romantic relationships. These three styles, similar to Ainsworths
initial classification of attachment, are 1) the secure style, which is typical of healthy
relationships in which people feel inclined to be available and reliable, and is
characterized by positive attitudes toward intimacy and cognitive flexibility; 2) the
anxious-ambivalent style, which is typical of those who express a mixture of anger
and disappointment as well as clinginess toward others, or demonstrate rigid or
hypervigilant behaviors; and 3) the avoidant style, which typifies those who tend to
avoid and appear emotionally detached from others, and may emphasize their self-
reliance as compensation for their reluctance to rely on others (Bowlby, 1988) (Hazan
& Shaver, 1994; Mikulincer & Florian, 1999). Attachment theory researchers in the
field of social cognition assume that attachment in adulthood exists primarily within
romantic relationships, and differences in the quality of those relationships determine
38


differences in attachment style (Stein, Jacobs, Ferguson, Allen, & Fonagy, 1998). It
has been suggested, however, that being married may itself influence attachment
style, and that attachment styles move toward security in more long-term or
committed relationships (Kobak & Hazan, 1991).
Mikulincer & Florian (1999) examined the association between spouses self-reports
of attachment styles and family cohesion among established married couples. Their
previous research found that people endorsing questionnaire items that indicated
insecure attachment styles (anxious-ambivalent or avoidant) reported more marital
conflict and dissatisfaction than those reporting secure attachment. Findings from
this study of 93 couples showed that secure attachment seemed to be related to
cohesive and flexible family relationships, whereas spouses who endorsed an avoidant
style reported relatively low levels of family cohesiveness and flexibility. Those
endorsing the anxious-ambivalent style reported high family cohesion but low
adaptability.
Attachment theory has not yet been investigated in the context of adult caregiving
relationships following the onset of disability. Several factors may complicate this
type of research. The post-SCI divorce rate may be as high as 48% (Kreuter, Sullivan,
Dahllof, & Siosteen, 1998), similar to that of the general population. In addition,
39


couples counseling, a seemingly important aspect of the initial rehabilitation
following injury, may be overshadowed by more immediately critical issues such as
home modifications, financing of equipment, and general adjustment to life with a
disability. Although it would be interesting to examine attachment style and marital
stability regardless of whether or not caregiving is an aspect of post-injury
relationships such a study needs to be conducted longitudinally over many years,
and therefore cannot be the focus of this investigation. Examining attachment
retrospectively among individuals who have separated or divorced following injury is
also difficult, as any assessment of pre-separation attachment is likely to be colored
by factors contributing to the marital demise. Studying attachment among couples
who remain together for many years following injury provides an important first step.
This investigation examines the degree to which attachment style mediates stress,
depression and burden for spouses of men with SCI. Similar attachment styles in
couples may be only one part of the picture, however.
Qualitative Aspects of Caregiving
While the studies described in the preceding discussion identified a number of
predictors and outcomes related to physical and psychological caregiver distress, few
have investigated the more qualitative aspects of caregiving. A qualitative research
approach can generate more exploratory (as opposed to confirmatory) information. In
40


addition, a qualitative approach takes advantage of the rich information provided by
those living the experience of marriage and spinal cord injury. It enables us to learn
what matters most to these caregivers, not what researchers believe matters. It can
also help explore the reasons for particular statistical relationships between variables.
Clearly, one challenge in assessing these qualitative issues involves recognizing the
high degree of subjectivity and difficulty measuring the impact of non-quantitative
factors on caregiver distress. Most quantitative measures of caregiver issues focus on
factors that are more negative stress, depression, and perceived burden. Few studies
have attempted to identity the positive side of caregiving, and even fewer have
incorporated qualitative methodologies to explore the experience of caregiving.
Schulz and colleagues (1988) reported that the negative effects of caregiving may be
counteracted in part by the benefits of caregiving. In their study, although caregivers
reported a great deal of distress associated with caregiving, they also reported a
number of beneficial effects. Almost 75% reported that caregiving made them feel
useful. Spouses, in particular, often viewed caregiving as contributing to their self-
worth. Caregiving may also improve relationships between caregivers and care
recipients and provide company for caregivers. However, this could be a tenuous
situation, if changes in the nature of close and loving relationships put the participants
41


at risk when one partner is required to assume greater responsibilities in the
caregiving role.
A qualitative study by Boykin and Winland-Brown (1995) exemplified the many
struggles and feelings experienced by caregivers the conflicts between love and
frustration. Respondents indicated that the love of their family member was the
impetus for the depth of the daily commitment to provide the care necessary moment
to moment, but caregivers often struggled with guilt as they tried to be genuinely with
and to retain respect and dignity o/loved ones. Although based on only a very small
sample, this study clearly illustrated both the positive and negative side of caregiving.
LeBlanc and colleagues (1997) suggested that studies should elaborate how
caregiving effects spill over to other domains of peoples lives, describing in more
depth the varied context in which stressful events and roles result in a proliferation of
stress. Zarit and colleagues (1986) found that caregivers learned to manage problems
more effectively, or they simply did not let problems bother them any more as time
went on. In addition, it has been suggested that pre-illness relationships between the
caregiver and care recipient may have an impact on the psychological impact of
providing care to an ill or disabled family member (Williamson, Shaffer, & Schulz,
1998). Studies to date have paid little attention to the prolonged or cumulative
42


consequences of being exposed to the stresses of caregiving over an extended period
of time.
Up to the present, much of the SCI caregiver research has focused solely on
quantifiable aspects of stress and depression in caregiving, leaving a considerable gap
in our understanding of the caregiving relationship in SCI. It has been suggested that
when an informant explains himself or herself by recounting a personal history, we
gain extraordinary access into that persons subjective experience (Ochberg, 1988, p.
173). This can be accomplished through some form of narrative. Narrative theory
postulates that people attempt to organize their experiences in the form of stories that
they see as true (Clark & Standard, 1997). Grant and colleagues (1998) have
indicated that
...there is strong evidence to suggest that family members talk about
caregiving not only as an extension of ordinary family relations and
everyday life, but also in terms of the purposes served by caregiving.
The purposes or ends to which caregivers direct their efforts can be
used as a powerful device to explain not only what caregivers do but
how they evaluate their experiences, (p. 60)
One specific form of narrative is the life story, which is a qualitative research method
for gathering information on the subjective essence of one person's entire life
(Atkinson, 1998). It begins as a recorded interview, is transcribed, and ends up as a
flowing narrative, completely in the words of the person telling the story. Through
43


the life story, information about an individual can be gathered on three levels of
personality description: personality traits, adaptations (including attachment styles),
and identity (McAdams, 1990). Details regarding the nature of relationships may also
emerge in the telling of life stories. With regard to the ways in which stories are told,
there is a finite number of basic story forms that people tend to adopt in creating
personal myths (McAdams, 1993, p. 50). These myths involve an imaginative
reconstruction of the past in light of an envisioned future (McAdams, 1993, p. 53).
The common characters present in these personal myths have been described by
McAdams (1993) as falling into categories of agency and communion, with agentic
characters described as those who act, think and feel in agentic ways (p. 134).
These individuals are described as those who are doers and who seek to control,
create, understand or act in other assertive ways. McAdams has classified these
individuals as the warrior, the traveler, the sage and the maker. The characters of
communion, on the other hand, are those who act, think and feel in communal ways
(p. 148). Their characteristics are more nurturing, passionate, or altruistic, and this
classification includes the lover, the caregiver, the friend and the ritualist. In
addition, McAdams suggests that there are individuals who are high in both agentic
and communal characteristics, and those who are low in both. The imago, or
character, types described by McAdams are shown in Figure 3.
44


According to McAdams, imagoes may personify aspects of who you believe you are
now, who you were, who you might be in the future, who you wish you were, or who
you fear you might become (p. 127). In telling ones life story, any one or a
combination of imagoes may be evident.
Figure 3
Common Characters in Personal Myth
High in Agency, High in Communion
The Healer
The Teacher
-
The Counselor ,y;:l Tif"- '''
The Humanist . -
The Arbiter .. ", -
, .;£> ;
Agency Communion
The Warrior . The Lover
The Traveler The Caregiver
The Sage The Friend
The Maker The Ritualist
Low in Agency, Low in Communion -
The Escapist
The Survivor

An interesting linkage of the life story interview with attachment theory was explored
by McAdams and colleagues (1997), who used Life Story Interviews in a study of 70
adults. The focus of the study was to determine how two groups individuals who
45


y
were more generative versus those who were less generative narrated their lives. In
this study, the authors identified a key theme of childhood attachment, based on
Ainsworths work, and assessed the extent to which attachments to mother, father,
siblings, grandparents, school friends and church friends were expressed. The
findings of these attachments were quantified to categorize each relationship into an
attachment style of anxious-ambivalent, avoidant, or secure. Of note was the fact that
there were no significant differences between the generative and non-generative
groups for childhood attachments, suggesting that highly generative adults do not
reconstruct their childhoods as any more secure in key interpersonal relationships than
do less generative adults (p. 687). It was clear from the narratives that a combination
of negative and positive factors or scenes played out in each of the life stories,
regardless of attachment or generativity, with positive outcomes often emerging from
more negative events.
The use of the Life Story Interview in SCI research is not unprecedented. Duggan and
Dijkers (1999) used this format to interview 40 individuals with SCI regarding their
quality of life. They noted that for most of their study participants, the SCI was
viewed as a serious disruption in the life story. Key themes that emerged included
elation after recovering something felt to have been lost (such as movement in the 2
2 G.enerativity refers to an adults concern for and commitment to the well-being of the next generation
(McAdams et al., 1997)
46


hands, a previously held job); affirmation that their lives had meaning for themselves
and others; accomplishments achieved; depictions of loss or threat of loss; issues
related to spirituality, and transformations in the way individuals saw themselves.
The value of narrative was felt beneficial to both clinicians working with individuals
with SCI, and to the individuals themselves, who experienced pleasure and
satisfaction in sharing their stories with others.
Research supports the concept that marital satisfaction is related to similarity of
marital attitudes (Plechaty, 1987), similar values and role expectations (Odell &
Quinn, 1998), greater congruence of what are believed to be important characteristics
in marriage (Fenell, 1993), similar perceptions of the spiritual character and
significance in marriage (Mahoney et al., 1999), and spousal personality congruence
(Richard, Wakefield Jr., & Lewak, 1990). Similarly, it could be hypothesized that
congruence in life story forms is related to greater marital satisfaction. In turn, this
increased satisfaction may mediate caregiver stress, depression and burden.
Therefore, it is anticipated that the rich nature of this narrative research can provide
greater understanding of the caregiving experience for spouses of people with SCI.
47


Specific Aims
The current study examines the multifaceted aspects of SCI caregiving in order to
gain greater understanding of this complex role and its relationship with health and
psychosocial outcomes. Using a mixed methods design (quantitative and qualitative),
this study investigates the extent to which perceived stress, depression, burden,
reported attachment style and congruence of life stories characterize the caregiving
experience in partners of men with SCI. To narrow the existing gaps in caregiver
research in SCI, this study addresses issues that have yet to be explored by pursuing
the following specific aims:
1. Examine caregiving issues in couples with SCI of long duration.
2. Compare differences in outcomes of stress, depression and burden between
caregiving spouses who provide fewer vs. those who provide more hours of
caregiving, and between caregiving spouses and their partners with SCI.
3. Explore the impact of attachment styles on caregiver outcomes.
4. Explore the impact of life story congruence between spouses on caregiver
outcomes.
Hypotheses
1. Partners of men with SCI will report greater a) stress; b) depression than their
injured partners.
48


2.
Partners of men with SCI who provide more than 10 hours per week3 of
assistance will report greater a) stress; b) depression; c) burden than those
providing 10 hours or less of assistance.
3. Older partners of men with SCI will report less a) stress; b) depression; c)
burden than younger partners.
4. In couples who demonstrate greater similarity of attachment styles, female
partners will report less a) stress; b) depression; c) burden than those in
couples with less similar attachment styles.
3 14 hours is a reasonable cut point based on expected functional outcome data from the Consortium for
Spinal Cord Medicine in their Clinical Practice Guidelines (Consortium for Spinal Cord Medicine,
1999). The need for assistance increases with higher neurologic levels of injury; the median amount of
assistance required for individuals with low level paraplegia is two hours per day. For this study, a
more limited amount of personal assistance (10 hours per week) is used, averaging the Functional
Independence Measure (FIMsm ) (Uniform Data System for Medical Rehabilitation,) expected amount
indicated for each neurologic level in the guidelines.
49


CHAPTER 3
STUDY DESIGN
This thesis utilized a non-experimental cross-sectional prospective cohort survey
design to gather quantitative information about caregiving and measures of stress,
depression, burden and attachment styles. A qualitative component of the study
involved in-depth interviews with caregivers and their spouses with SCI utilizing a
phenomenological approach. This allowed the participants to describe their lived
experiences as married couples in which the male partner has a SCI.
Participants
The study sample initially included couples in which the male partner had a spinal
cord injury for at least 10 years, and who had been together since prior to injury.
Excluded were individuals who were not in their current relationship prior to their
SCI or who had been injured less than 10 years. A ten year post-injury duration was
selected in order to examine the long-term impact of caregiving, as most previous
research in caregiving and SCI had focused on much shorter post-injury time periods.
50


Potential participants were identified from the Craig Hospital database. This database
contains detailed demographic and medical information on all individuals who have
been treated at Craig Hospital for rehabilitation and follow-up since 1976. Craig
Hospital is a subacute rehabilitation hospital that focuses its treatment programs on
people with spinal cord and brain injuries and conditions. Craig has a broad
catchment area for people with SCI that extends to all 50 states and some foreign
countries. In the 2002-2003 fiscal year 128 individuals were admitted for initial SCI
rehabilitation (consistent with previous years). Nearly half of these individuals were
discharged to Colorado residences, while the remaining individuals resided in other
states or countries. Individuals seen at Craig for rehabilitation come from a variety of
educational, vocational, ethnic and socioeconomic backgrounds. Those individuals
from Colorado are generally representative of the Colorado population demographic
profile.
Data regarding the marital status of all men with SCI who were at least 10 years post-
injury at the time of the study inception were extracted from the hospital database.
From this database exploration, 451 men met the criteria of being married at injury or
had unknown marital status at injury, with a presumed valid address or telephone
number. Addresses were presumed valid as long as the twice-yearly hospital
newsletter mailings (bulk mail) were not returned due to unknown addresses. In
51


reality, some individuals did not reside at the addresses stored in the hospital
database. This initial listing failed to yield the desired final sample size of 128 (based
on the. power calculations described below), so the inclusion criteria were expanded to
include couples in whom the male partner had a spinal cord injury for at least five
years, and who had been together since prior to their injury. This modification was
felt appropriate as it would still examine caregiving issues over a longer time post-
injury than most previous studies. This extension to 5 years post-injury yielded an
additional 82 individuals, for a total potential sample frame of 533.
Enrollment
All recruitment materials as well as an Informed Consent form were submitted to and
approved by the HealthONE Alliance Institutional Review Board (IRB), which
oversees research activities at Craig Hospital. On April 20, 2001, the IRB at the
University of Colorado at Denver accepted the protocol and approval granted by
HealthONE. After identifying potentially eligible individuals from the Craig
database, the investigator sent invitation letters to these individuals informing them of
the study, asking them to indicate whether or not they met the inclusion criteria (with
the same female partner since prior to injury), and whether or not they and their
partners were interested in participating. Reponses were requested via an eligibility
letter and these were returned in pre-stamped envelopes. No identifying information
52


about the respondents appeared on the outside of the envelopes, to protect their
identity. The study protocol required agreement to participate from both partners in a
couple. In cases where one or both partners declined participation, available database
information regarding the person with SCI provided information to determine if there
was response bias on the basis of age, time since injury and neurological level of
injury. Table 3 shows the results from the initial mailing to the 533 potential eligible
participants. It should be noted that concerted efforts to contact these individuals
were undertaken, including utilization of telephone operator assistance, electronic
telephone books, and verification of addresses from medical records.
Table 3
Potential Eligible Candidates
Participation Status N
Agreed to participate 137
Did not meet inclusion criteria 104
Did not wish to participate 111
Unable to contact* 178
Deceased 3
Total 533
*wrong telephone numbers; disconnected telephones; no answer at various
times of days, evenings and weekends, returned mail with no forwarding
address
53


After ascertaining study eligibility and willingness to participate from both partners,
the investigator sent a packet via mail containing a second letter explaining the study
methodology, two copies of an informed consent (one for each partner), the sets of
questionnaires (color coded with yellow forms to be completed by the partner with
SCI and green forms to be completed by the spouse), and a stamped return envelope.
The couples were asked to complete the questionnaires independently of one another
and not share their responses with one another until after completion, if they chose to
do so. Participants were asked to return the forms within three weeks of receiving
them. For those who did not return the materials, telephone follow-up occurred four
weeks after the mailing. At that time, potential participants were asked to complete
the materials within the next 14 days. For those who preferred, telephone interviews
were conducted separately with each member of the couple after they returned their
signed informed consent forms.
Power Analysis
Power and sample size calculations evaluated the ability to detect statistically
significant results. For all sample size calculations, the alpha level, the probability of
falsely rejecting the null hypothesis, was set at 0.05, while power, the probability of
correctly rejecting the null hypothesis, was set at 0.8. At a power of .8 and a
significance level of .05, a A standard deviation difference in group means for any
54


continuous normally distributed outcome variable could be detected for two equal
groups (spouse and partner with SCI; younger and older caregivers; wives providing
10 or fewer hours per week of assistance and those providing more than 10 hours per
week of assistance; similar or different attachment style) with 128 total subjects.
Thus, it appeared that the study would be sufficiently powered to examine the study
hypotheses.
Qualitative Sample
For the Life Story qualitative portion of this study, a convenience subsample of
couples was identified from the larger study sample of 137 participants. Although
convenience sampling is not the most desirable sampling strategy (Patton, 2002a), it
was felt that couples selected with this method would fairly represent the stories of
men with SCI and their wives. The convenience subsample was identified as those
couples who lived within the metropolitan Denver area. Nineteen men with SCI and
their spouses were identified from those who responded to the mail/telephone
questionnaire portion of the study as meeting this criterion. Of the 19 eligible
couples, nine agreed to participate in the Life Story Interviews (McAdams, 1995).
The ten couples who chose not to participate cited reasons of not having time to
engage in the interviews, not wishing to be audiotaped, or at least one of the partners
was not interested in participating in the lengthy interviews.
55


The consenting couples agreed to this segment of the study and signed separate
informed consent forms. The Life Story Interviews were conducted with each partner
separately and were audiotaped. The venue for these interviews was at the discretion
of the participants, and all occurred in their homes. Each life story interview lasted
from 90-120 minutes. Couples in this phase of the study were paid $25.00 each to
thank them for their participation.
Instruments (See Appendix A)
The instruments used in this investigation are described in detail below. Both the
individual with SCI and the partner were asked to complete the Center for
Epidemiological Studies Depression scale (CES-D) (Radloff, 1977), the Perceived
Stress Scale (PSS) (Cohen, Kamarck, & Mermelstein, 1983), and the Revised Adult
Attachment Scale (RAAS) (Collins & Read, 1990). In addition, the female partner
was asked to complete the Caregiver Burden Inventory (CBI) (Novak & Guest, 1989)
and a series of questions designed to collect demographic information (some of this
information was already available from the partner with SCI through the Craig
Hospital database). Based on a pilot test of the questionnaire packets with ten men
with SCI and their spouses, the investigator determined that completion of the
questionnaires took approximately 30-45 minutes. Upon completing and returning
56


the various questionnaires, each member of a couple was sent a $20.00 check to thank
them for their participation.
Center for Epidemiological Studies Depression Scale
(CES-D) (Radloff. 19771
The CES-D is a screening instrument used to measure the current level of depressive
symptomatology in general or clinical populations. The 20-item self-report scale asks
respondents to indicate the number of days during the past week they have
experienced symptoms such as feeling fearful, happy or lonely. This instrument
yields four subscale scores depressive affect, well-being, somatic symptoms and
interpersonal affect as well as a total score. For this study, only the total score was
calculated. The CES-D score ranges from 0-60, with a higher score indicating more
depressive symptomatology. The CES-D has been widely used within community
populations, including the elderly, to assess depressive affect and has been shown to
correlate significantly with clinical depression ratings (Radloff, 1977; Weissman et
ah, 1977; Roberts & Vernon, 1983; Radloff & Teri, 1986). Internal consistency
reliability assesses the consistency of results across items within a test, and for this
instrument is approximately .85 for the general population. The CES-D demonstrates
content, concurrent, construct and discriminant validity (Radloff, 1977; Radloff &
Teri, 1986). The investigator selected this instrument due to its broad use in the
caregiving literature, and previous use in samples of individuals with SCI and their
57


spouses (Weitzenkamp et al., 1997; Holicky & Charlifue, 1999). Both the individual
with SCI and the spouse responded to the CES-D in order to compare differences in
their scores.
Perceived Stress Scale (PSS~) (Cohen et al.. 1983)
This 14-item instrument measures the perceived stressfulness of various life situations
in community samples. Using five response options ranging from never to very
often, it asks subjects how often they have had particular thoughts or feelings during
the past month. Scores for the PSS can range from 0-56, with higher scores
indicating more stress. The PSS is reliable (.85) and correlates with life event scores,
depressive and physical symptomatology, utilization of health services and social
anxiety (Cohen et al., 1983). The investigator selected this instrument due to the
above reliability as well as its previous use in samples of individuals with SCI
(Weitzenkamp et al., 1997). Both the individuals with SCI and their spouses
completed this instrument in order to compare their scores.
Caregiver Burden Inventory (CBI) (Novak & Guest. 1989)
The CBI, a 24-item multi-dimensional instrument, measures the impact of burden on
caregivers. Analysis of the 24 items exploring their interrelationships reduced the
variables to five underlying factors: time-dependence burden, developmental burden,
58


physical burden, social burden and emotional burden. Responses to each item range
from 0 (not at all descriptive) to 3 (very descriptive). Alpha measures of internal
consistency range between .73 and .86 (Novak & Guest, 1989). Although originally
developed for caregivers of cognitively impaired individuals, the items appear to be
applicable for other types of disabilities.
Revised Adult Attachment Scale (RAAS)
(Collins & Read. 1990)
This 18-item instrument has been used to examine the relationship between adult
attachment styles and self-esteem, interpersonal attributions, social self-perception,
trust, and romantic love (Collins & Read, 1990). Respondents rate items on 5-point
Likert-type scales regarding their feelings about romantic relationships. While the
scale allows for a continuous assessment of attachment along three dimensions (trust
of others, feelings of anxiety in relationships and degree of comfort with closeness
and intimacy), it can also be scored to determine three types of attachment style
(Depend the extent to which subjects could trust others and depend on them to be
available when needed; Anxiety the extent to which subjects felt anxiety in
relationships such as feeling abandoned or not loved; and Close the extent to which
subjects were comfortable with closeness and intimacy). The RAAS has good
internal consistency (Depend = .75; Anxiety = .72 and Close = .69) and reliability
(Depend = .71; Anxiety = .52 and Close = .68) (Collins & Read, 1990.
59


Personal Support/Assistance Questionnaire
A team of health care and research professionals at Craig Hospital developed this
questionnaire to capture specific demographic and caregiving information about
individuals providing assistance to people with SCI. The investigator used this same
instrument in a previous study of people with SCI and their caregiving partners in the
United Kingdom (Weitzenkamp et ah, 1997).
In addition to questions about age, education level and relationship to the person with
SCI, questions ask about the time spent providing care and details of the types of
assistance provided. The instrument includes a 4-point Likert scale (l=poor; 2=fair;
3=good; 4=excellent) for which the responder indicates her perception of her own
current health and that of her partners current health. Responders also completed a
checklist of various physical and emotional consequences related to having a partner
with disability. This included items such as physical stress or strain, burnout,
isolation/loneliness, and fear for the future. A summation of the items checked
on this list of 18 issues provided a problems score. A higher score indicated more
perceived problems. Finally, respondents had the opportunity to consider if there was
anything they would change about their lives since their partners injury (yes/no), and
to elaborate what those changes would be. The dichotomous responses to the first
60


part of the question (no=0, yes=l) were used as predictors (independent variables) in
the regression analyses. Content analysis explored emerging themes arising from the
second portion of the question that described desired changes.
Life Story Interviews
A conceptual framework for this qualitative aspect of the study is McAdams life
story model of adult identity (McAdams, 1990). According to this model, people
living in modem societies begin to organize their lives in narrative terms in late
adolescence and young adulthood. People create internalized, evolving, and
integrative life stories that serve to reconstruct the past and anticipate the future in
such a way as to provide a life with a sense of unity and purpose. Like other literary
constructions, life stories can be analyzed in terms of plots, settings, scenes,
characters, and themes. According to McAdams (1993), key components of life
stories include nuclear episodes (high points, low points, turning points) (p. 259);
imagoes (idealized personifications of the self) (p. 122); and an ideological setting
(locates the personal myth within a particular ethical, religious, and epistemological
time and place, providing a context for the life story) (p. 81).
61


Analyses
Quantitative Analysis
Completed survey instruments were coded and entered into a Microsoft Access
database. Data were converted from Access files into an SPSS data file to perform all
quantitative analyses. Comparisons of key variables evaluated the degree to which
there were differences between the 137 couples who chose to participate and the 111
who were known to be eligible but did not wish to participate. Comparisons of
neurologic status of the men with SCI was done using chi-square tests one
comparing paraplegia versus tetraplegia (2x2), one comparing complete versus
incomplete neurologic status (2x2), and a third with individuals categorized on the
basis of functional and neurological status (3x4). These groups consisted of those
with functionally complete tetraplegia (TetraABC), those with functionally complete
paraplegia (ParaABC) and all those with functional preservation below the level of
the spinal cord injury (All D)4. Comparisons of age and time since injury used
independent samples t-tests.
4
Tetra ABC refers to those with functionally complete tetraplegia; Para ABC refers to people with
functionally complete paraplegia; and individuals with either functionally incomplete paraplegia or
tetraplegia are identified as All D, according to the International Standards for Neurological and
Functional Classification of Spinal Cord Injury (American Spinal Injury Association, 2002).
62


Caregiver ages, spouses time since injury, and CES-D, PSS, and CBI score group
means were reported using descriptive statistics. Next, in order to address each of the
study hypotheses, it was first necessary to compute the various scale scores, according
to each specific instruments specifications.
Hypothesis 1: Partners of men with SCI will report greater a) stress; b)
depression than their injured partners. To test this hypothesis, the CES-D and
PSS scores of the spouses with SCI and their caregiving partners were
compared using paired t-tests.
Hypothesis 2: Partners of men with SCI who provide more than 10 hours per week of
assistance will report greater a) stress; b) depression; c) burden than those providing
10 hours or less of assistance. The variable hours of care was used to categorize
the wives into two groups those providing 10 or fewer hours per week of assistance
and those providing more than 10 hours per week of assistance. This variable was
computed as follows:
(hours of care per weekday x 5) + (hours of care per weekend day x 2)
The sum of this calculation yielded the total number of hours per week of care.
Bivariate regression analyses with hours of care per week as the independent variable
were conducted. After this variable was dichotomized (provides <10 hours of care and
63


provides 10+ hours of care), differences in mean scores on the CES-D, PSS and CB1
were examined using t-tests. In addition, hours of care was used as a continuous
variable in the regression analyses described below.
Hypothesis 3: Older partners of men with SCI will report less a) stress; b) depression;
c) burden than younger partners. Bivariate regression analyses with age as the
independent variable were conducted. In addition, differences in mean scores
between wives who were less than 48 years old and those who were 48 or older were
examined using t-tests. Finally, age as a continuous variable was used in the
regression analyses described below.
Hypothesis 4: In couples where greater similarity of attachment styles is
demonstrated, female partners will report less a) stress; b) depression; c) burden than
those in couples with less similar attachment styles. The RAAS was analyzed with
chi-square tests to identify differences in attachment styles between spouses.
Comparisons of CES-D, PSS, and CBI scores based on same or different attachment
style were analyzed using t-tests.
Finally, linear regression analyses were performed in an attempt to discover what
variables were associated with worse scores on the main outcomes of stress,
64


depression and burden. For these analyses, each of the three main outcomes (stress,
depression, burden) was used as a concurrent predictor of other outcomes (e.g., PSS
score to predict CES-D or CBI score, CES-D score to predict PSS or CBI score, CBI
score to predict PSS or CES-D score), along with the independent variables: female
spouses age, time since partners injury, partners age, neurologic status of paraplegia
or tetraplegia (categorical variable), neurologic status of complete or incomplete
(categorical variable), hours of assistance provided (daily average time spent in all
caregiving tasks), partners PSS and CES-D scores, total number of items endorsed on
the physical and emotional consequences section of the caregiver questionnaire
(Question #16), self-perceived health, perceived health of the partner with SCI,
whether or not there was anything the female spouse would have changed, and
whether or not the attachment styles were the same or different. Most of the predictor
variables were selected due to the evidence in the literature that those variables were
related to outcomes of stress, depression or burden. The three outcomes of interest
and their predictor datasets are shown in Table 4, indicating the primary study
measures which were used as either dependent or independent variables in the
specific analyses.
All independent variables were first correlated with each outcome to identify bivariate
significant correlations. Those significant variables were then entered into a
65


regression model for each outcome (stress, depression, burden) using multiple linear
regression. Beta coefficients (standardized regression coefficients) were calculated.
Betas further from 0 (either positively or negatively) represent a stronger association
between the predictor and the outcome. Positive values of a beta represent a positive
relationship between the predictor and the outcome, where an increase in the predictor
value is associated with an increase in the outcome variable value. Conversely,
negative values of a beta represent a negative relationship between the predictor and
the outcome.
Table 4
Description of Linear Regression Analyses
Outcome Independent variables
Stress (PSS), Depression (CES-D) Burden (CBI) Female spouses age, time since injury, partners age, neurologic status (paraplegia/tetraplegia), neurologic status (complete/incomplete), hours of assistance provided, total number of physical/emotional consequences items endorsed, self-perceived health, perceived health of partner with SCI, whether or not would have changed the past, same or different attachment style
Stress Above, plus spouses and own CES-D score, spouses PSS
(PSS) score, CBI total and subscale scores.
Depression Above, plus spouses and own PSS score, spouses CES-D
(CES-D) score, CBI total and subscale scores.
Burden Above, plus spouses and own CES-D score, spouses and
(CBI) own PSS score.
66


Qualitative Analysis
It should be noted that both internal and external validity checks were incorporated
into the qualitative analyses. Internal validity is a measure of the trustworthiness of
the findings, however, in qualitative research, validity can be conceptualized as
credibility. The credibility of the qualitative data in this study rested on the
employment of as much triangulation between the coders as possible (given time,
scope, and budgetary constraints) regarding the emerging themes and concepts.
Responses from the Personal Support/Assistance Questionnaire to the question
Looking back on your life, and knowing what you do now, is there anything in your
life since your partners injury that you would change? were analyzed in order to
identify any common or recurring themes. Some of these themes were determined a
priori but additional emerging themes were also examined. Responses were coded by
the investigator and two independent research assistants (analyst triangulation). The
coders met to compare their individual findings and develop a set of themes. These
were then categorized by the primary investigator, and these categories were
confirmed with the two research assistants.
67


Analysis of the audiotaped Life Story interviews used content analysis. According to
Patton (2002b), content analysis is used to refer to any qualitative reduction and
sense-making effort that takes a volume of qualitative material and attempts to
identify core consistencies and meanings. The interviews were transcribed in
Microsoft Word and imported into the Atlas.ti software package for coding and
analysis. The investigator and two independent research assistants scrutinized the
data separately, identifying recurring themes or concepts. At the end of this initial
process, the three compared their findings. Discrepancies were discussed and, even if
mutual agreement was not reached, those themes or concepts were included in the
listing. This listing also included all of the consistent themes shared by the three
analysts. The primary investigator then re-coded each interview in Atlas.ti with the
identified code system. This process was confirmed by the research assistants.
Finally, the new or possible explanatory connections between categories/themes and
subcategories were identified using axial coding. In this process, the numerous codes
were reassembled into larger categories or themes that emerged from the actual
interviews as well as a priori themes identified from prior work by McAdams and
colleagues (1997). Specific themes of self-evaluation, self-efficacy (Bauer &
Bonanno, 2001), agency and communion (McAdams, 2001) were incorporated into
5 Agency includes concepts of strength, power, expansion, mastery, control, dominance, autonomy,
separation and independence.
6 Communion includes concepts such as love, friendship, intimacy, sharing, belonging, affiliation,
merger, union and nurturance.
68


the coding scheme. Each theme could have both a positive and negative component.
For example, a theme of family support would include phrases relating to the
beneficial aspects of having supportive relatives (positive family support), as well as
text describing a lack of such support (negative family support).
Although the sample for the qualitative interviews was quite small, these themes were
finally used to assign individuals into specific imagoes (based on McAdams
conceptualization, described on page 45 of this thesis). In addition, caregiving
spouses and their partners with SCI were compared with each other to determine if
there were thematic differences in their life stories. Specific similarities or differences
with regard to peak events, low points, turning points, important/influential people,
and spiritual/religious belief systems were compared.
69


CHAPTER 4
QUANTITATIVE RESULTS
This chapter presents the quantitative analyses of study data and tests each of the
previously presented study hypotheses. Based on the findings and supported by the
literature, additional analyses were conducted to explore possible relationships
between various demographic and other independent variables and the selected study
outcomes of stress, depression, and burden. The results of the analyses are described
in detail below.
Sample Bias Analyses
In order to determine if the sample of individuals who participated in this study were
different from those who were eligible but chose not to participate, chi-square tests
comparing paraplegia vs. tetraplegia, complete vs. incomplete neurologic status, and
neurologic group were performed. Age and time since injury were compared using
independent samples t-tests. Results are shown in Table 5.
As indicated in the table, there were no significant differences in age, duration of
injury, or neurologic status between those who participated and those who did not.
70


Study participants tended to have paraplegia, incomplete injuries, ox functionally
complete paraplegia. They were also younger and injured for a slightly shorter period
of time. Therefore, it appeared that individuals who participated in the study did not
differ from those who did not.
Table 5
Comparison of Characteristics of Individuals Included
and Not Included in the Study
Characteristics In Study Not In Study Sig.
Number Percent Number Percent
Paraplegia or Tetraplegia -
Paraplegia 74 54.0% 51 45.9% .128
Tetraplegia 63 46.0% 60 54.1%
Complete or incomplete * V*
Complete 64 46.7% 55 49.5% .376
Incomplete 73 53.3% 56 50.5%
Neurologic group ' v'
Tetra ABC 46 33.6% 45 40.5%
Para ABC 62 45.3% 45 40.5% .527
All D 29 21.2% 21 18.9%
Mean age 51.2 (sd 11.51) , 53.9 (sc 12.51) .079
Mean duration of injury 11.71 (sd 5.28) 11.44 (sd 5.25) .692
71


Caregiver Study Sample Characteristics
Table 6 presents the demographic characteristics of the wives of men with SCI in the
sample.
Table 6
Demographic Characteristics of Female Spouses in the Study
Demographic Characteristics Number Percent
Education
Less than high school 1 1%
High school/GED 44 32%
Some college 60 44%
College or nursing degree 13 10%
Graduate degree 19 14%
Mean age ; 49.3 years (sd 11.5; range 25-79)
Mean length of time knew spouse 28.9 years (sd 12.6; range 7-75)
Mean length of time providing care 9.9 years (sd 6.6; range 0-31)
Note: Percentages rounded to the nearest 1%.
This sample of spouses was relatively well-educated, with most having attended
college for some period of time. They were also, on average, approximately 2 years
younger than their partners with SCI. Although the mean duration of injury, as shown
in Table 5, was nearly 12 years, the mean length of time providing care was slightly
shorter.
72


Tests of Study Hypotheses
Hypothesis 1: Partners of men with SCI will report greater
a) stress; b) depression than their injured partners
Weitzenkamp and colleagues (1997), in a study of British individuals with SCI and
their spouses, found depression to be greater among the spouses than in their injured
partners, but there were no significant differences in stress scores. The current study
explored these relationships in a different population using the same assessment
instruments. Scores on the PSS and CES-D were compared for the men with SCI and
their partners using paired samples t-tests. Individuals with any missing data were
excluded from each analysis. Table 7 shows the results of the paired t-test analysis.
Similar to the British study, there were no significant differences in PSS scores, but
there was a statistically significant difference in CES-D scores; however, not in the
expected direction. In this sample, the men with SCI had significantly higher scores
on the CES-D than their partners. It is notable that the current study sample was
younger and injured, on average, a shorter time, compared to the British sample,
which may account for the contradictory findings on the CES-D.
73


Table 7
Comparison of PSS and CES-D Scores Between Men with SCI
and their Partners
Men with SCI Female Partners N Sig.
Mean s.d. Mean s.d.
PSS 21.47 8.15 21.57 8.61 123 .906
CES-D 15.43 12.16 12.35 9.95 105 .018
Hypothesis 2: Partners of men with SCI who provide more
than 10 hours per week of assistance will report greater
a) stress; b) depression; c) burden than those providing
10 hours or less of assistance
There is evidence suggesting that caregivers who provide more assistance (due to
severity of disability or other related factors) may be at greater risk for developing
stress, depression or subjective feelings of burden (Zarit et al., 1986; Gaynor, 1990;
Miller et ah, 1995; Leblanc et ah, 1997; Weitzenkamp et ah, 1997). To evaluate this,
bivariate linear regression analyses of scores on the PSS, CES-D and CBI, with hours
of care provided per week as the independent variable, were performed. Results of
these analyses are shown in Table 8. Both the CES-D and PSS scores were
significantly related to hours of care, but the CBI total score was not. It is notable that
the distribution of hours of care was positively skewed, with 24 of the spouses
74


reporting they provided no hours of care per week. In addition, the CBI total scores
were also positively skewed, with 60 spouses reporting no burden.
Table 8
Bivariate Regression Analysis of Hours Per Week Caregiving
With CES-D, PSS and CBI Scores
P Sig.
CES-D Score .192 .044
PSS Score .187 .042
CBI Total Score .098 .276
CES-D, PSS and CBI total and individual subscale scores were then compared for the
spouses providing an average of 10 or fewer hours per week of assistance to those
providing more than 10 hours per week of assistance to their spouses with SCI (the
rationale for selecting 10 hours as a cut point is provided on page 50). The analysis
was done using an independent samples t-test with hours of care as the factor. Results
are shown in Table 9.
When dichotomizing hours of care per week, there were no statistically significant
differences in either the PSS or CES-D scores between the two groups, however, there
was a trend toward greater reported stress and more depressive symptomatology
among the spouses providing more than 10 hours of assistance per week to their
partners with SCI. The CBI total score and all subscales, with the exception of
75


Emotional Burden, were significantly different. Those spouses providing more than
10 hours of assistance to their partners reported greater time, developmental, physical
and social burden than those providing 10 or fewer hours per week of assistance.
There was, however, no significant trend of greater reported emotional burden among
spouses providing more assistance. It is possible that the selected cut-point of 10
hours per week influenced these findings.
Table 9
Comparison of PSS, CES-D and CBI Scores by
Hours Per Week of Assistance Provided to Spouse with SCI
<10 Hours/Week >10 Hours/Week .
Mean Sd N Mean Sd N Sig.
PSS 19.8 7.4 58 22.5 9.1 65 .233
CES-D 10.5 9.1 56 13.9 9.9 56 .959
CBI-Time 2.2 2.8 60 4.9 5.0 69 .000
CBI-Development 1.9 3.2 60 3.8 4.6 70 .000
CBI-Physical 1.1 2.2 60 3.4 4.5 69 .000
CBI-Social 1.0 2.1 60 1.9 3.3 69 .001
CBI-Emotional 0.6 1.5 60 0.8 1.5 70 .258
CBI (Total) 6.6 9.9 60 13.9 15.6 67 .000
76


Hypothesis 3: Older partners of men with SCI will report
less a) stress; b) depression; c) burden than younger partners
Previous studies have suggested that younger caregivers of people with Alzheimers
disease fare worse in terms of psychosocial outcomes than do older caregivers (Fitting
et ah, 1986; Zarit et al., 1980). In order to determine if this phenomenon is true in
SCI, caregiver age was explored to test the above hypothesis. First, bivariate
regressions of stress, depression and burden, with age as the independent variable,
were performed. These analyses yielded no statistically significant results, as shown
in Table 10.
Table 10
Bivariate Regression Analysis of Female Spouses Age
With CES-D, PSS and CBI Scores
P Sig.
CES-D Score .064 .491
PSS Score .086 .338
CBI Total Score .120 .169
The ages of partners were then dichotomized into two groups; those less than or equal
to 48 years old and those greater than or equal to 49 years old. The cutoff age of 48
represented the median age of the spouses. The analysis used the independent
samples t-test with caregiver age as the factor. Results of this analysis are shown in
Table 11.
77


Table 11
Comparison of PSS, CES-D and CBI Scores by Two Age Groups
<48 Mean Years C Sd Hd N >49 Mean Years C sd )ld N Sig.
PSS 20.9 7.7 63 22.3 9.6 64 .019
CES-D 11.7 10.1 60 13.2. 9.8 58 .925
CBI-Time 2.9 4.0 69 4.3 4.5 66 .084
CBI-Development 2.5 3.8 69 3.2 4.3 67 .069
CBI-Physical 1.9 3.3 69 2.6 4.1 66 .076
CBI-Social 1.2 2.3 69 1.7 3.3 66 .017
CBI-Emotional 0.6 1.4 69 0.8 1.7 67 .224
CBI (Total) 9.0 13.2 69 11.7 14.0 64 .411
In the analysis by two age groups, the PSS scores differed significantly, but not in the
hypothesized direction, with the older spouses exhibiting greater stress than the
younger spouses. The only other significant difference was in the CBI Social Burden
subscale, but again, older spouses reported greater burden, and this finding was not in
the hypothesized direction. There were non-significant trends toward the older
spouses exhibiting greater depressive symptomatology and greater CBI total, time,
developmental, physical and emotional burden scores.
78


Hypothesis 4: In couples where greater similarity of
attachment styles is demonstrated, female partners will
report less a) stress; b) depression; c) burden than those
in couples with less similar attachment styles
Subjects with SCI and their spouses were assigned to the attachment style category in
which they scored highest. In the event that the subjects with SCI and their spouses
highest scores were the same for two different categories, those two groups described
the attachment style. For example, the three key categories for this instrument are
anxious, close, and dependent. In cases with tied high scores, the additional
categories were anxious/close, anxious/dependent, and close/dependent.
Table 12 indicates the numbers of pairs in each of the above categories, as well as the
numbers of men with SCI and the numbers of wives in each category. Data were
missing for one of the wives, so the couple was excluded from this analysis. A couple
was considered matched if a combined style, such as anxious/close matched
either an anxious or close style in the partner.
79


Table 12
Attachment Styles of Individuals and Couples
Attachment Style N (%) Men with SCI N (%) Wives N (%) Pairs
Anxious 7 (5%) 4 (3%) 1 (<1%)
Close 75 (55%) 86 (63%) 46 (34%)
Dependent 39 (29%) 36 (27%) 12 (9%)
Anxious/Close 2 (2%) 2 (2%) 4 (3%)
Anxious/Dependent 1 (>1%) 0 (0%) 0 (0%)
Close/Dependent 13 (10%) 8 (6%) 18(13%)
Unmatched - ; 'My- 'p , 55 (41%)
Note: Percentages rounded to the nearest 1%.
Clearly, the majority of both men and women were categorized as having a close
attachment style, followed by dependent styles. Within relationships, the greatest
percentage of couples did not have a similar attachment style, but for those with a
similar style, most couples were categorized as close. This finding may be explained
by the long-term nature of these relationships post-SCl.
A dichotomous variable was then created based upon whether or not the subjects with
SCI and their spouses attachment styles matched (0=no, l=yes). Using an
independent samples t-test with attachment style as the factor, the results are shown in
Table 13.
80


Table 13
Comparison of PSS, CES-D and CBI Scores Among Female Partners
by Similarity of Attachment Style
Attac Same iment Styles Different Attachment S tvles Sig.
Mean Sd N Mean Sd N
PSS 22.0 9.0 56 21.3 8.5 71 .746.
CES-D 12.7 10.3 54 12.2 9.7 64 .586
CBI-Time 3.5 4.0 59 3.7 4.6 76 .132
CBI-Development 2.8 3.8 59 2.9 4.2 77 .212
CBI-Physical 2.5 3.8 59 2.0 3.6 76 .439
CBI-Social 1.5 2.7 59 1.5 2.9 76 .539
CBI-Emotional 0.9 1.8 59 0.6 1.2 77 .019
CBI (Total) 11.1 13.7 59 9.7 13.6 74 .595
In couples where greater similarity of attachment styles was demonstrated, female
partners did not report statistically significantly less stress, depression or burden than
those in couples with non-similar attachment styles. The only significant finding was
for the emotional burden subscale score of the CBI; however, this was not in the
hypothesized direction. There were non-significant, seemingly counter-intuitive
trends toward greater depressive symptomatology, greater stress, more physical
burden and more social burden in the female spouses whose attachment style matched
that Of their spouses with SCI.
81


Regression Analyses
Regression analyses identified combinations of variables that were associated with
depression, stress, and burden scores. Variables selected for entry into the regression
analyses were initially chosen based on findings from the literature suggesting
relationships between specific independent variables and the outcome variables.
Within each regression analysis, a subset of these variables was selected if the
bivariate correlation, as shown in Table 14, was significantly correlated with the
selected outcome (PSS, CES-D, CBI total and subscale scores) at the .10 level. For
example, higher CES-D scores were significantly correlated with more hours per
week spent in caregiving, more items endorsed as being a problem (summation of
problems), poorer health of the partner with SCI, poorer health of the female spouse, a
positive response to would have changed the past, a higher score by the spouse with
SCI on the CES-D, a higher PSS score for both the spouse with SCI and the female
partner, a higher CBI total score, and higher CBI subscale scores with the exception
of the time dependence burden subscale score.
82


Table 14
Bivariate Pearson Correlation Coefficients for Selected Independent
and Outcome Variables
Independent Variables CES-D Score PSS Score CBI Total Score
Time since injury .218
Age of partner with SCI .160
Complete (1)/Incomplete(0) .153
Hours/week caregiver .192 .187
Sum of problems .350 .401 .463
Health of partner with SCI -.354 -.347 -.159
Health of female partner -.531 -.468 -.171
Would change past .205 .315
CES-D of female partner - .759 .245
CES-D of male partner .307 .327 .164
PSS of female partner .759 .282
PSS of male partner .309 .409
CB1 total .245 .282
CB1 time dependence burden .851
CBI developmental burden .236 .295 .919
CBI physical burden .227 .308 .868
CBI social burden .296 .370 .816
CBI emotional burden .204 .251 .728
Table 15 shows the results of the linear regression analysis for the CES-D. In this
model, the female spouses self-perception of health and her PSS score had
significant associations with CES-D scores. A better score of self-perceived health
was related to less depressive symptomatology, as evidenced by a lower CES-D score,
whereas higher (worse) PSS score was associated with greater depressive
symptomatology.
83


Table 15
Regression Analysis for Female Spouses CES-D Scores
Independent Variables P t Sig. Adj. R2
Total hours per week caregiving -.042 -0.542 .590
Summation of problems .134 1.285 .203
Health of partner with SCI .031 0.333 .740
Health of female partner -.313 -3.511 .001
Would change past -.117 -1.403 .164
Male partners CES-D score .218 1.552 .125
Female partners PSS score .605 6.742 .000 .534
Male partners PSS score -.185 -1.333 .187
CBI total score .590 1.450 .151
CBI Developmental burden -.246 -1.157 .251
CBI Physical burden -.354 -1.776 .080
CBI Social burden -.119 -0.744 .459
CBI Emotional burden .040 0.319 .751
The above regression was also performed using all independent variables, not just
those that were significant in the bivariate correlation. This analysis resulted in the
same model and adjusted R2. Missing values were then imputed using mean
substitution, and regression analysis with variables having significant bivariate
correlations was again performed. This analysis yielded a model that included the
same independent variables as that without mean substitution. All further regressions
for outcome variables were thus performed without mean substitution.
84


The results of the regression analysis for the PSS scores are shown in Table 16. In
this model, a higher PSS score for the male partner and a higher score on the CES-D
for the female spouse were significantly associated with greater stress.
Table 16
Regression Analysis for Female Spouses PSS Scores
Independent Variables P t Sig. Ad j. R2
Total hours per week caregiving -.044 -0.561 .576
Summation of problems .053 0.504 .616
Health of partner with SCI -.064 -0.692 .491
Health of female partner .006 0.059 .953
Would change past .136 1.634 .106
Male partners CES-D-score -.197 -1.397 .166
Female partners CES-D score .604 6.742 .000 .535
Male partners PSS score .322 2.384 .020
CBI total score -.209 -0.508 .613
CBI Developmental burden .088 0.409 .683
CBI Physical burden .150 0.738 .463
CBI Social burden .049 0.308 .759
CBI Emotional burden -.022 -0.175 .862
For the CBI regression analysis, significantly correlated CBI subscale scores were
excluded from the analyses in order to allow other variables to enter into the model.
Table 17 shows the results of that analysis.
85


Full Text

PAGE 1

PARTNERS OF MEN WITH SPINAL CORD INJURY: AN EXPLORATION OF STRESS, DEPRESSION AND BURDEN by Susan B Charlifue B.A., University ofDenver, 1976 M.A., University ofDenver, 1986 A thesis submitted to the University of Colorado at Denver in partial fulfillment of the requirements for the degree of Doctor of Philosophy Health and Behavioral Sciences 2004

PAGE 2

2004 by Susan B. Charlifue All rights reserved.

PAGE 3

This thesis for the Doctor of Philosophy degree by Susan B. Charlifue has been approved by David P. Tracer Deborah S. Main Gale G. Whiteneck Date

PAGE 4

Charlifue, Susan B. (Ph.D., Health and Behavioral Sciences) Partners of Men with Spinal Cord Injury: An Exploration of Stress, Depression and Burden Thesis directed by Associate Professor David P. Tracer ABSTRACT A spinal cord injury (SCI) can have a devastating affect on both the individual sustaining the injury and family members. Within an overarching conceptual framework of attachment theory, a sample of 13 7 couples who were married before the SCI and had remained married a minimum of 5 years post-injury were interviewed to explore quantitative and qualitative perceptions of stress, depression and burden in wives of men with SCI. Stress was measured by the Perceived Stress Scale (PSS), depression by the Center for Epidemiological Studies Depression Scale and burden by the Caregiver Burden Inventory (CBI). Nine couples participated in additional qualitative interviews. The study hypotheses were: 1) partners of men with SCI will report greater stress and depression than their injured partners; 2) partners of men with SCI who provide more than 1 0 hours per week of assistance will report greater stress, depression and burden than those providing 1 0 hour or less of assistance; 3) older partners of men with SCI will report less stress, depression and burden than younger partners; and 4) in couples who demonstrate greater similarity of attachment styles, female partners will report less stress, depression and burden than in those couples with less similar attachment styles. Results indicate that the men with SCI were significantly more depressed than their wives. Wives who provided more than 10 hours of assistance had significantly greater perceived burden than those who provided 1 0 or fewer hours of assistance. Older spouses had significantly higher PSS scores than younger spouses. The only significant difference related to attachment style was in the Emotional Burden subscale of the CBI, which was higher for those in relationships with more similar attachment styles. The qualitative interviews highlighted the importance of establishing independence for husbands and wives, and the importance of family and of social support on helping minimize perceptions of stress, depression and burden. IV

PAGE 5

This investigation has provided new and unique insights into the nature of marriages that survive SCI offering information to health and service providers to guide them in developing intervention programs that may help families cope more effectively with the consequences of SCI. This abstract accurately represents the content of the candidate's thesis. I recommend its publication David P. Tracer v

PAGE 6

DEDICATION I dedicate this thesis first and foremost to my husband, Gerry, for his support, encouragement, and patience. He willingly took on additional responsibilities within our household that I, too often, neglected while completing this degree. His love and faith in me never flagged during this process, for which I am extremely grateful. I also dedicate this to my parents, Miriam and Ralph Wexler, who instilled in me a spirit of independence curiosity, and desire to succeed. Their confidence in my abilities has served as a gentle driving force throughout my life. Finally, I dedicate this to those individuals with spinal cord injuries and their partners who shared their stories with me. They have taught me how resolute the human spirit can be, and how adversity can be addressed, acknowledged, and overcome. I have drawn strength from their strength.

PAGE 7

ACKNOWLEDGMENTS I would like to acknowledge the outstanding support and help from my thesis committee chair, Dr. David P. Tracer, and from my committee members, Dr. Lucinda Bryant, Dr. Ann Gonzales, Dr. Deborah Main and Dr. Gale Whiteneck. They each provided expertise and insight, as well as continued encouragement in helping me prepare this thesis. I would also like to acknowledge my colleagues in the Research Department of Craig Hospital, particularly Kenneth Gerhart, MS, RPT; Hang Pham-Bowman, MA, Kendra Noble, BA, Chris Cusick, MA, Cynthia Harrison-Felix, PhD, and Amy Dannels, ND RN. All were instrumental in providing support, helping with the data collection, providing insight and suggestions to facilitate the data analysis, and generally acting as a cheering squad during this process. This project was supported by grant number H133000001 from the National Institute on Disability and Rehabilitation Research, U.S. Department of Education Office of Special Education and Rehabilitative Services.

PAGE 8

CONTENTS Figures ............. ............................................. ................... ................. ....... xii Tables ......................................................................................................... xiii CHAPTER 1. INTRODUCTION ......................................................................................... Rationale for Conducting the Study .................................................... 2 Arrangement of the Thesis .................................................................... 4 2. REVIEW OF THE LITERATURE ............................................................... 6 Introduction ..... .... ..................................................................... ............ 6 The Nature of Caregiving ...................................................................... 8 Caregiver Studies with Non-SCI Samples .......................................... 14 Burden ........................................................................................ 14 Stress ................................. ................................ ....................... 18 Caregiver Health and Depression ..................... ... ....................... 26 Caregiver Studies with SCI Samples ..................................... ... ........... 30 Attachment Theory ............................ .............................. .................. 36 Qualitative Aspects of Caregiving ....................... ............................... 40 Specific Aims ................ .................... ... .... ............................ ............ 48 Vlll

PAGE 9

Hypotheses ..................................... .................................................... 48 3. STUDY DESIGN ........................................................................................ 50 Participants .......................................................................................... 50 Enrollment ............................... .... .... ....... ............................................. 52 Power Analysis .................................................................................... 54 Qualitative Sample .............................................................................. 55 Instruments ................................................ ......... ............................ ... 56 Center for Epidemiological Studies Depression Scale (CES-D). 57 Perceived Stress Scale (PSS) ....................................................... 58 Caregiver Burden Inventory (CBI) ............................................... 58 Revised Adult Attachment Scale (RAAS) ................................... 59 Personal Support/ Assistance Questionnaire ................................ 60 Life Story Interviews .................................................................... 61 Analyses .............................................................................................. 62 Quantitative Analysis ................................................................... 62 Qualitative Analysis ..................................................................... 67 4. QUANTITATIVE RESULTS ..................................................................... 70 Sample Bias Analyses ............................................. ... ......................... 70 Caregiver Study Sample Characteristics ................ .... ......................... 72 IX

PAGE 10

Tests of Study Hypotheses ................. .......... ............... ..................... 73 Hypothesis 1 : Partners of men with SCI will report greater a) stress; b) depression than their injured partners .......... 73 Hypothesis 2: Partners of men with SCI who provide more than 1 0 hours per week of assistance will report greater a) stress; b) depression; c) burden than those providing 10 hours or less of assistance ....................................... 74 Hypothesis 3: Older partners of men with SCI will report less a) stress; b) depression; c) burden than younger partners ..... ......................... ...... ................................... 77 Hypothesis 4: In couples where greater similarity of attachment styles is demonstrated, female partners will report less a) stress; b) depression; c) burden than those in couples with less similar attachment styles .................... 79 Regression Analyses .................................................. ......................... 82 5. QUALITATIVE RESULTS ........................................................................ 92 Personal Support/ Assistance Questionnaire ........................................ 92 Life Story Interviews .... ................................ ..................................... 97 General Themes ............ ............................................. ... .......... 99 Critical Events .......... :: .................................................. ............. 109 Significant People ...................................................................... 120 Future Script .................................................................... .......... 124 Stresses and Problems ......................... ......................... ............ l27 Personal Ideology ....................................................................... 129 X

PAGE 11

6. DISCUSSION .... ........ ........................................ ..... ......... ; .................... 141 7. CONCLUSIONS ....................................................................................... 152 APPENDIX A. DATA COLLECTION INSTRUMENTS ................................... 157 B. IRB APPROVAL LETTERS .................. ................. ................. 173 REFERENCES .......................................................... ........................ ...... .......... 180 XI

PAGE 12

FIGURES Figure Alzheimer's Caregiver Stress Model .......................................................... 19 2 Stress Appraisal Model ............................................................................... 22 3 Common Characters in Personal Myth ....................................................... 45 4 Proposed SCI Caregiver Stress Model.. .................................................... 149 XII

PAGE 13

TABLES Table 1 Summary of Relevant Literature-Non-SCI Studies .................................. 11 2 Summary of Relevant LiteratureSCI Studies ..................... .... .... ............ 13 3 Potential Eligible Candidates ....... ................................. ...... ............. ....... 53 4 Description of Linear Regression Analyses ................................................ 66 5 Comparison of Characteristics of Individuals Included and Not Included in the Study ......................... ............................ .............. 71 6 Demographic Characteristics of Female Spouses in the Study ........... ..... 72 7 Comparison ofPSS and CES-D Scores Between Men with SCI and their Partners ..................................................................................... ... 7 4 8 Bivariate Regression Analysis of Hours Per Week Caregiving with CES-D, PSS and CBI Scores .............................................................. 75 9 Comparison ofPSS, CES-D and CBJ Scores by Hours Per Week of Assistance Provided to Spouse with SCI.. .............................................. 76 10 Bivariate Regression Analysis of Female Spouse s Age with CES-D, PSS and CBI Scores .............................................................. 77 11 Comparison ofPSS, CES-D and CBI Scores by Two Age Groups ............ 78 12 Attachment Styles oflndividuals and Couples .................. ............... ....... 80 13 Comparison ofPSS, CES-D and CBI Scores Among Female Partners by Similarity of Attachment Style ................................... ............ 81 Xlll

PAGE 14

14 Bivariate Pearson Correlation Coefficients for Selected Predictor and Outcome Variables ............................................................................... 83 15 Regression Analysis for Female Spouses' CESD Scores .......................... 84 16 Regression Analysis for Female Spouses' PSS Scores ............................... 85 17 Regression Analysis for CBI Total Scores .................................................. 86 18 Regression Analysis for CBI Time Dependence Burden ............................ 87 19 Regression Analysis for CBI Developmental Burden ................................. 88 20 Regression Analysis for CBI Physical Burden ............................................ 89 21 Regression Analysis for CBI Social Burden ............................................... 89 22 Regression Analysis for CBI Emotional Burden ................................ : ....... 90 23 Themes Regarding What Wives Would Have Changed ............................. 93 24 Description of Qualitative Sample .............................................................. 98 xiv

PAGE 15

CHAPTER 1 INTRODUCTION Serious physical trauma or illness, while undoubtedly devastating to the individual incurring such an event, can also have a far-reaching impact on family members and close friends. This is particularly true when the trauma results in limitations that necessitate physical assistance from others. Spinal cord injury (SCI) is one such trauma that often results in dependence of the injured party on other people. As such, it provides an opportunity to evaluate the broader consequences of disability on family and friends; in particular, the impact of such a disability on a marital relationship. Given the young age at which most of these injuries occur, it is not surprising that more than half of individuals are single at the time of injury, and approximately 31% are married (the remainder are divorced, separated or widowed at the time of injury) (National Spinal Cord Injury Statistical Center, 2003). Divorce following SCI is relatively frequent, with a rate at least as high as the nearly 50% noted in the general population (De Vivo & Fine, 1985; Kreuter, 2000). The percentage of individuals divorced continues to rise in the years following injury (National Spinal Cord Injury Statistical Center, 2003). In the general population evidence suggests married individuals have lower risk of mortality than those who are divorced, widowed or never married (Smith & Zick, 1994; Johnson, Backlund, Sorlie,

PAGE 16

& Loveless, 2000; Kravdal, 2001; Lin, Rogot, Johnson, Sorlie, & Arias, 2003). Being married also results in a lower risk of various poor health outcomes, including depression (Caimey, Boyle, Offord, & Racine, 2003), injury (Hokby, Reimers, & Laflamme, 2003), obesity and cardiovascular disease (Hedblad et al., 2002). The beneficial impact of marriage on life satisfaction for people with SCI also has been demonstrated (Post, de Witte, van Asbeck, van Dijk, & Schrijvers, 1998; Holicky & Charlifue, 1999; Dowler, Richards, Putzke, Gordon, & Tate, 2001 ). Therefore, it can be reasonably postulated that maintaining marriages post-injury may contribute to long-term well-being for the person with SCI. A question remains, though, whether the spouse shares this benefit. A spouse who has a partner with SCI may instead experience depression or stress related to the injury and its other negative sequellae. A spouse who provides personal assistance may feel some sense of burden as well. However, it is also possible that a spouse may experience a sense of fulfillment, usefulness or other positive benefits of assuming a caregiving role for a partner with a severely disabling condition. The long-term consequences of SCI on the partner, therefore, warrant greater investigation. Rationale for Conducting the Study Because SCI usually occurs in young male adults, this population provides an ideal group among whom to study the impact of long-term caregiving within marriage. A 2

PAGE 17

significant body of research, which is described in detail in Chapter 2, indicates that caregiving spouses have high risk for stress, depression, experiencing a sense of burden and even physical deterioration. While this existing research is plentiful very little focuses on spouses of individuals with SCI and almost none addresses these issues in the years following the injury. Therefore, this study explores perceptions of stress depression and burden in female spouses of men with SCI who were married prior to the injury and have remained married for at least 5 years post-injury. Attachment theory a conceptual framework that describes how individuals make strong affectional bonds to others (Bowlby, 1977), provides a theoretical basis within which to examine the nature of the relationship between partners to determine whether or not similarity in attachment style affects caregiver stress, depression or burden. This study explores the impact of SCI on female spouses, whether or not they provide personal assistance to their injured partners. More specifically, this study helps address the following questions: 1. Do wives of men with SCI report greater stress and depression than their injured partners? 3

PAGE 18

2. Do wives of men with SCI who provide more than 10 hours of assistance per week of assistance report greater stress depression and burden than those providing 10 hours or less of assistance? 3 Do older wives of men with SCI report less stress, depression and burden than younger wives? 4. Does greater congruence in attachment in couples contribute to less reported stress, depression and burden by wives? The findings from this study are important for two reasons: 1) they will remediate a large gap in the literature which to date, has not focused on the long-term consequences of caregiving in SCI, and 2) they will help guide rehabilitation professionals who work with couples following SCI to help identify potential areas of risk for marital discord and build on marital strengths early in the post-injury phase. / An-angement of the Thesis This thesis comprises seven chapters Chapter One provides a briefbackground and rationale for conducting this study. Chapter Two reviews the theoretical and empirical background for this study describing the known physical consequences of spinal cord injury the emotional impact of caregiving and providing the background of attachment theory which guided this study. 4

PAGE 19

Chapter Three describes the study design and methodology, including descriptions of the instrumentation, sampling and analysis procedures conducted. Chapter Four gives the demographic details of the sample and results of quantitative analyses of the collected data. Chapter Five reports findings from qualitative analyses of interviews conducted with a subsample of couples and discusses the themes and patterns that emerged from the interviews. Chapter Six provides an overview and discussion of the quantitative findings, enhanced by information from the qualitative data, and compares these findings with the existing research, and describes the limitations of the study. The thesis concludes with Chapter Seven, which discusses the potential clinical implications of these findings and directions for future research. 5

PAGE 20

CHAPTER2 REVIEW OF THE LITERATURE I am his lover, his wife, his friend, his financial partner, his cheerleader, his business partner, his housekeeper, and at times his tormenter. My feelings for him run deep -I love him like he is a part of my soul. [He] is my cheerleader, my positive thinker, my mechanical mind while I am his hands, my partner, my scapegoat for my sad or anxious feelings. (wife of man with spinal cord injury) Introduction SCI typically causes immediate and usually permanent manifestations: an alteration or loss of muscle function and sensation in the body below the level of the injury, as well as impairment in bowel and bladder emptying and sexual function. These conditions alone mean that the injured individual often requires assistance from others to carry out day-to-day functions such as using the toilet, dressing, bathing, or getting out of bed. Other body systems also are likely to be affected. For example, the respiratory system in people with injuries of the upper thoracic and cervical spine is often compromised due to the lack of adequate musculature to produce a strong cough or inhale deeply. People with high cervical injuries in whom the diaphragm is partially or wholly paralyzed often require fullor part-time ventilator support. In addition, the absence 6

PAGE 21

or alteration of sensation may produce problems with skin breakdown due to pressure over a bony prominence or skin trauma from performing certain mobility activities such as transferring from the wheelchair to the cat or toilet. Osteoporosis due to paralysis and disuse is commonly felt to be an underlying risk factor for pathologic fractures following spinal cord injury. Lower extremity osteoporosis develops rapidly in the first post-injury year, with a loss of about one-third of the original bone mass by 16 months post-injury, increasing the likelihood of fractures (Garland et al., 1992). -1:' Just as spinal cord injury can have a major effect on organ system functioning in the acute injury phase, evolving post-injury physiology leaves SCI survivors with autonomic and somatic nervous system dysfunction, which results in lasting impairment of many organ systems. The cumulative effect of this pathophysioiogy over many years of survival may result in the development of secondary complications, which will become increasingly prevalent with a longer duration of impairment (Charlifue & Lammertse, 2002). These conditions may, in tum, have an impact on the ability to remain independent in a variety of activities, and assistance from others becomes critical. Because the raison d'etre of clinicians and professionals involved in SCI rehabilitation is to reduce disabiifty the independence with which individuals perform activities (Ditunno, Cohen, Formal, & Whiteneck, 1995), enabling them to live in the community (Dijkers, Buda Abela, 7

PAGE 23

Gans, & Gordon, 1995), individuals optimally will receive needed caregiving in the home. In fact, over 91% of all people with SCI are discharged to home settings after initial inpatient acute care and rehabilitation (Eastwood, Hagglund, Ragnarsson, Gordon, & Marino, 1999). 1 Evidence suggests that in the years following rehabilitation, people with SCI will have an increasing need for help as they age (Gerhart, Bergstrom, Charlifue, Menter, & Whiteneck, 1993). This, coupled with the fact that 40% of all people with SCI are now over the age of 45 and one-fourth have been injured for 20 or more years (Menter, 1993), makes it clear that the number of people who require assistance will only increase in the coming years. The Nature of Caregiving In the United States, most home care is provided through informal channels by family members (Mintz, 1996). Similarly, almost 70% of people with SCI-the majority of them men receive some form of assistance and support from family members (Nosek, 1990). In 1985, Schulz and Decker reported that 41% of people with SCI named only one support person, most frequently a female spouse. One individual, therefore, very frequently assumes most or all of the responsibilities of caring for a 1Depending on factors such as the neurological severity of injury, funding, age, equipment provision or concomitant complications, the length of inpatient rehabilitation hospitalization can vary; however, the 2002 reported mean length of stay within the SCl Model Systems ranged between 7-36 days for acute care and 18-73 days for rehabilitation (National Spinal Cord Injury Statistical Center, 2003). 8

PAGE 24

person with a disability (Decker, Schultz, & Wood, 1989). This responsibility carries with it several physical and emotional risks. Pearlin and colleagues (1990) have identified a caregiving model that measures constructs such as overload (the caregiver's sense of being overwhelmed by tasks); loss of seif(the degree to which the caregiver feels a personal loss of his/her own identity during caregiving); management of meaning (the degree to which respondents are able to find meaning in their situations); and expressive support (the degree of help and support from friends and relatives. They describe caregivingas the behavioral expression of one's commitment to the welfare of another, whereas caring is the affective component. Under this definition, all caring relationships include some degree of assisting one another. However, as the authors indicate, when serious or prolonged impairment figures into the relationship, caregiving may become the dominant component of the interactions and may, under some circumstances, "expand to the point where it occupies virtually the entirety of the relationship" (Pearl in et al., 1990, p. 583). Consistent with the dominance of caregiving over other life experiences, at least some of those who provide assistance to loved ones report diminished social activities, less personal time, lack of privacy, and increased anger and depression (Given, Collins, & Given, 1988). 9

PAGE 25

The most commonly studied themes of caregiving focus on stress, depression, and burden, which are discussed in detail below. The vast majority of research regarding spouse caregiving has been done in the area of Alzheimer's disease, with a primary focus on caregiver stress. However, as interest and awareness of caregiver needs increase, investigators have started looking at the multifactorial nature of caregiver distress and caregiver life satisfaction. The following review of the literature discusses caregiving and disability in this broader context, particularly examining the emotional and social facets of caregiving. Caregiving research in spinal cord injury is described separately, as many of the issues of caregiving (and hence, the ways in which they can be studied) likely differ. Specifically, this review examines several studies and describes the basis of each on underlying theories of stress, depression, or burden. Tables 1 and 2 review the breadth of findings in the literature. 10

PAGE 26

Table 1 Summary of Relevant LiteratureNon-SCI Studies ndy Information Study Sample Findings Limitations .. <[ ., .: ... <' :. :: ; .... ::. : .:;, : .,-_.: .. .... .:: .>: ..... ,_.;";"";;'.:--. :'b.:; rnor 1990 125 female spouses of men with Extended time in caregiving contributed to more perceived Reports of stress based on comments from e Control Design neurological diagnoses; control burden, more illness episodes participants; no attempt to tie qualitative and group of 30 with healthy spouses quantitative findings nsch & Spaid 131 caregivers (70% female; 89% Men and women reported different types of fonnal services Small proportion of males, mixed diagnoses mal 9 white); mean age 69 years; average used; men more likely to receive home help Women more generalizations difficult; coping measured in week training length of marriage 39 years. 51% likely to participate in outside activities use counseling and absence of any stress assessment; possible bias for elderly of care receivers had dementia; other services; few gender differences in use of informal to recruitment strategy (volunteers identified caregivers; preothers included multiple diagnoses support. Women reported more coping problems and more through self-referral or agency-referral) rvention data limiting activities of daily living subjective burden than men Memory and behavior problems key Qredictors if et al 1986 64 spouses (33 female; 31 male) of Compared burden, social support, quality of relationship in Factors other than caregiver burden may have le J year follow-up people with Alzheimer's disease; couples with partner living at home vs living in institution. to placement decision ; multiple !-tests in analysi n earlier study; in-mean ages 72.27 male and 63 37 Burden lower for wives at follow-up compared to initially; may be limitation small sample size; variety of 1e interviews female; generally well-educated, also lower for women who had placed spouse in nursing factors may have contributed to change in financially comfortable home. Social support not a factor in placement decision responses between time I and 2 'i;'{i#i'' :;r;,;' ,, ,,. ,-: >. ::1t,, n:-:.:;f< -.---03W'' :f;' ill[: ;5i]s.: '''' .2:::;;,. lcr et all995 215 spouses ofpeople with Role s train (stress) and depression highly correlated Comparatively small proportion of African a from a study of Alzheimer's disease (64% female, Stressor variables included spouses' limitations in physical American males; difficult to interpret cultural 11ence of race and 64% white); mean age 74.7 years ADLs, caregiver reactions to disruptive behavior, and differences; sample is quite elderly and findings der on spouse caregiver self-reported health. Depression predicted by the may not be generalizable to younger caregivers above plus lower sense of control and lower appraisal of caregiving confidence. m s et al 1992 181 primary caregivers of recently Caregivers exhibited more psychological symptoms on brief No theoretical framework on which to base worl jy to as sess hospitalized veterans who required symptom inventory compared to population norms; reported not truly a study of stress per se, but an assessm' itive, neutral and 2+ hours of assistance. Mean family functioning levels similar to normal population; ofstressors, with family functioning as outcome ative aspects of caregiver age 50.5; mean months of caregiver adjustment related to family communication and Biased toward experiences of female caregivers care provision 37.5. problem solving caregiver subjective health and length of time providingcare

PAGE 27

s s s tressor s and rds of c aregiving ; caregivers :itudinal study :ht et al 1998 cip a nts in survey omen's health s compared with carers ; attempt to r ate stress burden et all998 : s process model edict physical and :al healthi tudinal Table 1 (continued) Summary of Relevant Literature Non-SCI Studies 95 daught e rs, daughters-in-law of parents needin g help with at least one of three activities of daily living ; care g ivers also occup y ing roles of mother and wife Me a n age 43.9 y e ars, 94% white 642 informal AIDS caregivers ; mean age 40 years ; 368 with longitudinal follow-up, 28% caregivers seropositive at time 1, 67% positive at time 2 122 70% female, 60 % white, 49% spouses 30% daughters 21% other relative s Mean caregiver age 59.57 y ears Rank ordering of stressors and rewards for each role. Stress in at least one role accounted for significant portions of varianc e in well-being scores. Role stress score s relat e d to poorer well being and reward s cores related to better well-being. Higher stress in caregiver role alone related to better well-being than for those with stress in all three roles. Reports of poor health increased as assistance with activities increased Care-related stress related to physical health of caregiver ; care recipient problem behaviors, caregiver work strain and unemployment associated with poorer caregiver health Role overload significant+ correlate of poor health in seropositive caregivers. Seronegative males report better health than seronegative females. Caregiver depression a correlate of health Small sample size selection bias (volunteers), limite< to daughters so results could not be generalized to caregiving spouses. N e ed to e xami ne role quality nc just role occupancy Due to unique nature of AIDS, high proportion of seropositive caregivers and relatively young age of caregivers, results difficult to generalize to other samples of caregivers pant over to assess outcomes in caregivers who had lost or had to place care recipient; mixture of caregiver relationships. Study of dementi a patients only-can't generalize to physical impairments

PAGE 28

Table 2 Summary of Relevant Literature SCI Studies Indy Information Study Sample Findings Limitations tker et a11989 67 primary caregivers; 88% of Caregivers and people with SCI scored similarly on SJTiall samples size; no assessment of stress. Burde dy to i dentity caregivers female, mean age 53.6 measure of life satisfaction but slightly lower than assessed by single-item question, not by validated enninants of life years ; 85% spouses of person with general adult public. No significant differences in instrument. Mixture of spouses and other family sfaction and SCI depression scores between caregivers and general caregivers. ressio n in primary population. Caregivers with better perception of own egivers of middlehealth are less depressed and have better life d and elderly satisfaction. Greater levels of social support, greater pie with SCI sense of control over their lives, and less perceived burden associated with higher levels of life satisfaction and less depression. 'itzenkamp et al 124 spouses of people with longSpouses scored significantly higher than partner with No assessment of marital strength or marital 17 term SCI ; 90% female, mean age 52 SCI on measures of depression. Spouses scored higher adjustment; mixture of preand post-injury tish longitudinal years; mean length of marriage 29 on mea su res of stress. No difference in well-being marriages ; small sample size; British sample makes jy of aging with years, 52% married pre-injury 64% scores Significantly higher proportion ofcaregiving generalization to US caregiving samples difficult I perform some caregiving activities spouses reported physical stress emotional stress, burnout, fatigue, anger and resentment compared to noncaregiving spouses. Actual scores on measure of perceived stress did not differ. Caregiving spouses had significantly higher depression scores. !wchuk ct all998 62 caregivers for people with recent Physical symptoms of caregivers positively related to No mention of relationship between person with SC 1 dy of caregiver SCI ; 87% female, mean age 40.1 initial status of depressive behavior. Initial level of and caregiver (spouse? parent?) ; small sample size; comes in first year years, assessed during positive affect inversely related to depressive behavior. caregiving assessed early in injury course-no ;t-SCI hospitalization, at I, 6 and 12 Rate of change in depress ive behavior associated with indication of long-term changes/characteristics; stre months post-discharge expressive support Anxiety was best predictor of not measured objectively as a construct inferred frc variability in initial levels of physical symptoms. measure of anxiety. Physical symptoms statistically significant in predicting variability in anxiety and depressive behavior Younger patient age associated with more anxiety at initial measurement and low positive affect associated with higher initial depression scores.

PAGE 29

Caregiver Studies with Non-SCI Samples Disability, whether its onset is slow and progressive or sudden, affects not only the individual but the family as well, particularly when they are called upon to provide help to their loved one. The degree of impact, however, may vary. The literature does not provide a dear conceptualization of what constitutes caregiver distress. In several studies, investigators have focused on caregiving stress and depression, and their association with a number of factors such as the caregiver's sense of control (Miller et al., 1995), family functioning (Evans et al., 1992), and the severity of disability (Zarit et al. 1986). Another component of the caregiver role is the evaluation of burden, or role strain. While depression, burden and stress may be closely related, researchers often treat them as separate constructs and assess them with different measurement tools. The following review looks at these various types of caregiver distress. Burden Zarit and colleagues (1986) defined burden as the extent to which caregivers perceive their emotional or physical health, social life and financial status as suffering as a result of caring for their relatives. Burden includes many factors outside the caregiving role, such as family commitments employment, and other demands that 14

PAGE 30

compete for the caregiver's time and energy. Bush and Job (1993) reported that nearly one fourth of the caregivers in the US work the equivalent of one and one half jobs when employment and caregiving responsibilities are combined Even in the absence of paid employment, informal caregivers take on additional tasks not normally associated with their familial or spousal roles. Caregivers often must provide home health care assume responsibility for most or all of the household chores, and become the primary wage earner, the primary or only parent, and the healthcare gatekeeper (Holicky, 1996). Caregivers may find that these increased responsibilities leave little time for their own needs According to Bush and Job (1993), compensating for the emotional drain of constant responsibility is one example of invisible (unobservable) work. This type of activity is difficult to quantify, but its presence can undoubtedly create added burdens on the caregiver. Other examples of such unobservable work include resolving guilt about negative feelings toward the person to whom assistance is provided, coping with upsetting behavior of the dependent person working through role reversal in the lifelong relationship between caregiver and care recipient and satisfying the need for creativity and originality to offset tedious routines. Gaynor (1990), in a study of 155 caregivers, reported that women who had been involved in caregiving for an extended period of time had greater feelings of perceived burden, and the length of time one 15

PAGE 31

had engaged in home caregiving was associated with an increase in illness episodes and/or perceived burden in the caregiver. Not only is burden theoretically viewed as a factor in caregiver strain, it is measured with several instruments. Barusch and Spaid (1989) studied 131 caregivers (92 women) providing assistance to family members with various diagnoses, including dementia, stroke, lung disease and general frailty. Caregivers responded to items that measured the number of specific tasks the caregiver performed for the care recipient and the method of coping with caregiver problems. An inventory that presented 34 situations that might pose problems for the caregiver, scored using a five point Likert scale of coping effectiveness, assessed caregiver problems and coping. Caregiver burden was measured using the Zarit et al 20-item burden scale (Zarit, Reever, & Bach-Peterson, 1980). Women reported more coping problems than did men, but few differences in coping styles were identified. In addition, women caregivers reported experiencing significantly higher subjective burden than did men (p S001), with the most important predictive variable being the number of memory or behavior problems exhibited by the care recipient. The second most important factor, caregiver age, decreased burden. That is, younger caregivers reported greater perceived burden. 16

PAGE 32

Zarit and colleagues ( 1986) reported the amount of burden caregivers felt depended on the degree to which they coped well or poorly with specific problems related to the care recipient's disabilities and the resources they had available, especially for obtaining assistance and occasional relief from care giving tasks. They also pointed out that "while there has been little theory to guide research, it is frequently assumed that the caregiver s burden is strongly related to the severity of the patient's disabilities" (p. 260). In this two-year longitudinal investigation of 64 caregiving spouses they found that subjective factors (such as perceived emotional or physical health, social life and perceived suffering as a result of providing care) on the part of the caregiver were more important than the severity of the care recipient's illness in determining burden. The authors concluded that their findings regarding burden were consistent with general buffering models of stress which propose that the impact of harmful events is mediated by whether subjects perceive them as harmful and by their own coping responses (Lazarus & Folkman, 1984b). The variety of tools used to assess burden often differ substantially in their conceptualization ofburden, with some instruments providing a simple total score that is intended to reflect burden (Zarit, Reever, et al. 1980) and others using a multidimensional approach incorporating component subscales to assess different aspects ofburden (Novak & Guest, 1989). Furthermore, the majority of validated 17

PAGE 33

burden assessments have been developed for and tested with samples of people providing assistance to individuals with cognitive disabilities. In order to evaluate burden as an outcome in other disability groups, therefore, the multidimensional measures may show more promise, as the cognitive component scores can be interpreted separately or, if appropriate, disregarded. Clearly, studies of care giving highlight the variety of factors that may contribute to perceived burden, including physical status of the care recipient, duration of disability, multiple role responsibilities, and availability of social support systems. In spite of potential buffers and rewards of caregiving, however, the burden literature is replete with findings associating perceived burden with increased stress, depression, and other negative outcomes. It is critical, then, to consider the interrelationship of these other outcomes and their possible predictors in caregiver research. Stress Stress is a condition that results when an individual has a perceived discrepancy between the demands of a situation and the resources of that person's biological, psychological, or social systems (Sarafino, 1990). 18

PAGE 34

Caregiver stress refers to ongoing problems within the caregiving role that cause disruptions or changes that have potential to tax or exceed the caregiver's resources (O'Brien, 1993). Pearlin and colleagues have conceptualized a stress process model of Alzheimer's caregivers, illustrated in Figure 1. Figure 1 Alzheimer's Caregiver Stress Model I > ; Mediators This model comprises four domains, the first of which is the background and context of distress, which includes factors such as socioeconomic characteristics, caregiving history, and composition of the family and network. The second domain, the stressors, includes primary stressors, which encompass the demands of care giving, (problematic behavior, dependencies in daily activities) and the perceived hardships 19

PAGE 35

experienced by the caregiver that can manifest as fatigue and overload. Also included in the second domain are the secondary stressors of role strains (family conflicts, work/caregiving conflicts, economic problems, and social life limitations), and intrapsychic strains (self-esteem, mastery, role captivity and competence). The third domain involves mediators of stress, such as coping skills and social support. Finally, the fourth domain represents the outcomes of stress, which include depression, anxiety and difficulties with physical health. While conceptually useful, this model only suggests a one-way influence of mediators on the stressors-it does not address the impact of the various stressors on possibly breaking down the coping capabilities or social support network of the caregiver. In addition to the stress process model, stress and coping models provide a focus for much of the caregiving stress research. In these models, combinations of stressors (e.g., perceived burden) and resources (e.g., available social support or finances) influence an individual's response to stress (Lazarus & Folkman, 1984a). This model is illustrated in Figure 2. In this example, a stressor is evaluated in terms of its significance or threat (primary appraisal). If the threat is considered minimal or if the consequences of the "stressor" are potentially positive, the individual would not perceive this stressor as a threat. In the process of secondary appraisal, individuals evaluate their ability to control or change the situation, manage their own reactions 20

PAGE 36

and evaluate their coping resources. If the resources or sense of control are adequate, the threat is reduced. These appraisal processes, then, contribute to one's perception of the stressor. The entire process is fluid, and consists of frequent re-appraisal of the stressors and resources. Miller and colleagues (1995) used the stress and coping theoretical framework when examining issues of control as they relate to caregiver distress. In their study of 215 spouses of people with Alzheimer's disease, they looked at predictors of caregiver depression or role strain. Stressor variables included the spouse's limitations in physical activities of daily living (P ADLs) such as dressing or grooming, task distress (the amount of effort or difficulty experienced by caregivers in the performance of caregiving-related tasks), behavior problem upset (caregiver reactions to disruptive behavior, such as emotional lability or aggressive behavior), and the caregiver's selfreport of physical health. To assess psychological resources, the investigators examined the caregiver's sense of control, based on the Mirowsky and Ross conceptualization (Mirowsky & Ross, 1990). This measure compares claims of responsibility for good or bad outcomes (instrumental attributions) with claims denying responsibility for good or bad outcomes (fatalistic attributions), and assesses one's sense of personal responsibility for one's own life and sense of control over others. 21

PAGE 37

Figure 2 Stress Appraisal Model (based on Lazarus & Folkman 1984a) PRIMARY APPRAISAL *consequences *harm Positive Consequences Little/No Harm Not a threat SECONDARY APPRAISAL *resources *ability to cope Adequate Resources Good Ability to Cope Threat reduced Finally, in this same study, caregiver mastery-one's general feeling of competence in caregiving-was examined (Miller et al., 1995). Significant predictors of depression included poorer perceived health, greater behavior problem upset, higher levels of PADL impairments in the spouse and greater task distress, a lower sense of control and lower appraisal of competence in care giving. They also found high positive correlations between role strain and depression scores. In addition, they reported that a sense of control and caregiver mastery moderated the effects of stressors on depression. A highlight of this particular investigation was the multifactorial approach to caregiver distress; however, they did not attempt to assess the impact of 22

PAGE 38

other variables that could shed light on factors mitigating caregiver stress, such as quality of the caregiving relationship. Evans and colleagues ( 1992), in their investigation of caregiver stress, did address issues of relationships, but did not identify a particular theoretical framework on which to base their work. Rather, they illustrated some of the gaps in existing research-most notably, the focus on the negative aspects of caregiving. In their work, they evaluated 181 caregivers for psychological adjustment and family functioning. Their findings supported the notion that family functioning, described partly as communication and problem solving, reliably predicted caregiver adjustment. Gender also appears to play a role in the perceived degree of caregiver distress, with a predominant theme of greater depression among caregiving wives than caregiving husbands (Alspaugh, Stephens, Townsend, Zarit, & Greene, 1999; Beeson, Harton Deutsch, Farran, & Neundorfer, 2000; Neundorfer et al., 2001; Beeson, 2003). In addition, wives who provide assistance to their husbands with a disability report greater depression than non-caregiving spouses (Beeson, 2003). A population-based study of 1 002 male and female informal caregivers found that women were more likely than men to experience greater emotional challenges associated with caregiving 23

PAGE 39

and attempting to balance other family and work obligations (Navaie-Waliser, Spriggs, & Feldman, 2002). Women in this study were more likely than men caregivers to deliver more intensive care such as providing assistance with bathing, dressing, and managing incontinence. In an effort to examine both the negative and potentially positive aspects of caregiving, Parris Stephens and colleagues (1994) examined role stress and rewards as predictors of well-being. They studied 95 women simultaneously occupying the roles of caregiver, mother and wife, looking at both the competing-demands hypothesis (Brody, 1981) and the expansion hypothesis (Marks, 1977). The former hypothesis assumes that the roles people (specifically women) occupy in addition to that of caregiver compete for their time and energy, and such responsibilities make them more vulnerable to stress. Conversely, the latter posits that energy gains, rather than energy expenditures, are accrued by individuals with multiple roles, predicting positive consequences of multiple roles. The investigators measured stress and reward in three arenas of the women's lives-caregiver, mother, and wife. A single self-rating item measured the physical health of the respondents. Role overload was also assessed with a single item; rated on a five point Likert-type scale. Stress in at least one role accounted for significant portions of variance in well-being scores. Role rewards explained additional variance and were significant predictors of well24

PAGE 40

being even when controlling for role stress. In addition, women experiencing higher levels of stress in their caregiver role alone demonstrated better well-being than did the women experiencing larger amounts of stress in all three roles. It is unfortunate that this study's limitations included small san1ple size, selection bias of the sample, and a cross-sectional design that precluded interpretations of any causal links between role stress, role rewards and well-being. It did, however, provide a valuable contribution to our understanding of the impact of multiple roles on caregiver well being, and highlighted positive factors related to the caregiving role in its relationship to other life roles. The above studies, like those in the burden literature, demonstrate the multidimensional nature of caregiver stress. The concept of buffers or mediating factors is more clearly illustrated using the stress process models. However, the measurement of stress as a unique construct is not consistent, and is often an assumed outcome based on more objective measures of burden, depression, physical health of the caregiver, and tasks performed. Rarely is stress, itself, measured. Rather, it is interpreted from measures of anxiety, burden depression, or other manifestations of distress. In order to evaluate stress as a specific outcome and to associate it with other outcomes and predictors, a measurement tool designed to address stress specifically appears needed. 25

PAGE 41

Caregiver Health and Depression A pmiicular concern for caregivers is the risk of injury incurred while performing caregiving tasks. Numerous studies have identified caregiving as a risk factor for health problems, including more episodes of infectious illnesses (Kiecolt-Glaser, Dura, Speicher, Trask, & Glaser, 1991 ), poorer immune responses to vaccines for flu and pneumonia (Kiecolt-Glaser, Glaser, Gravenstein, Malarkey, & Sheridan, 1996; Glaser, Sheridan, Malarkey, MacCallum, & Kiecolt-Glaser, 2000; Vedhara et al., 1999), delayed or slowed wound healing (Kiecolt-Glaser, Marucha, Malarkey, Mercado, & Glaser, 1995), development of mild hypertension (Shaw et al., 1999) (Grant et al., 2002); coronary heart disease (Vitaliano et al., 2002) and overproduction of the proinflammatory cytokine, IL-6 (Kiecolt-Glaser et al., 2003). Schultz et al. (1990) reported that it could reasonably be predicted that caregivers should have higher rates of physical illness, since they are chronically exposed to high levels of stress. Even in the absence of actual bodily injury or illness, caregiving may negatively affect perceived health, often manifested as role strain, stress or depression. Symptoms and health problems related to caregiving include hypertension, arthritis, back pain, stomach upset or indigestion, and cardiac problems (Bush & Job, 1993) and a generalized decline in perceived health (Douglas & Daly, 2003). Furthermore, Gaynor (1990) reported that caregivers tend to ignore their own 26

PAGE 42

health and are more apt to forego health promoting behaviors if those behaviors conflict with caregiving, and thus end up in a cycle of neglect. LeBlanc and colleagues (1997) investigated the effect of stress on health among caregivers of persons with AIDS using the stress proliferation framework ofPearlin and associates (Pearlin, 1989; Pearlin et al., 1990). In this model, primary stressors are defined as difficulties associated with the actual tasks of caregiving, and stem directly from the needs of the care recipient and the nature and magnitude of the care demanded by these needs. Secondary stressors, on the other hand, are problems that arise as a consequence of providing care, but do not directly entail the tasks of caregiving. These may include factors such as family and job conflicts, financial problems, or loss of self-esteem. The secondary stressors are generated and exacerbated by the primary demands of caregiving in a process of stress proliferation. The LeBlanc study asked a sample of 642 infonnal caregivers of persons with AIDS to self-assess their frequency of experiencing symptoms of eight common ailments (headaches, cold/chills/sore throat, indigestion, constipation, diarrhea, weakness, backaches, shortness ofbreath). Six items from an established symptoms checklist assessed depression. In addition, an array of care-related stressors were measured, including perceived stressors arising from the provision of help, stressors related to personal financial strain and conflicting demands between work and caregiving. As 27

PAGE 43

assistance with activities increased, so did reports of poor health. Caregiver depression was also a significant correlate of poor physical health. In contrast, Knight and colleagues (1998) surveyed 742 women, ofwhom 90 were engaged in caregiving for a family member with a traumatic brain injury. They found no significant elevation in subjective ratings of depression comparing the caregivers to those who did not report having a caregiving role, suggesting that caregiving is not necessarily deleterious to health and well-being. Goode and associates ( 1998) also used a stress process model to describe the role of primary stressors on caregiver health, positing that a number of processes can occur over time to explain changes in caregiver mental and physical health outcomes. To assess the interplay of numerous factors, they examined changes in primary caregiving stressors, appraisal, coping, social support and mental and physical health outcomes over a one year period. Their 122 participants reported caregiving stressors as they related to the care recipient's impairments and need for assistance. As the authors expected, changes in appraisal of stressfulness were positively related to changes in depression and physical health symptoms, and changes in depression were positively correlated with changes in health symptoms. These findings supported the usefulness of a stress process model, demonstrating that psychosocial resource 28

PAGE 44

variables (appraisal, coping responses, and social support) predicted longitudinal changes in caregiver mental and physical health. A significant consequence of caregivers' neglect of their own physical and emotional health is their increased risk of becoming injured or too ill to provide assistance to their dependent partners. Their own perceived poor health, in tum, may lead them to report more depression and less life satisfaction (Decker et al., 1989), and render them less able to cope with their various responsibilities. When caregivers are unable to cope effectively with role responsibilities, the health and well-being of the care recipient also may be jeopardized (Elliott, Shewchuk, & Richards, 1999). Ultimately, the inability to continue providing care because of declining physical or health may lead to institutionalization of the care recipient (Burton, Newsom, Schulz, Hirsch, & German, 1997). Institutionalization may be considered an acceptable option for frail elderly individuals who have cognitive impairments, such as with Alzheimer's disease. Individuals with spinal cord injury, who may be younger and still have many years of life, however, may find it less desirable, both for themselves and for their caregiving spouses. The literature regarding physical injury, illness and depression is almost inextricably interwoven with reports of caregiver stress and burden. However, in the non-SCI 29

PAGE 45

caregiver literature, depression is rarely measured separately. Again, in order to fully evaluate depression and its unique relationship to other caregiver outcomes, as well as the factors which may be predictive of caregiver depression, it would need to be measured with a well-validated tool. A significant contribution to the caregiver literature, therefore would be to treat these distinct outcomes both independently and in relation to one another, and use well-validated instruments designed to evaluate the varying aspects of caregiver burden, stress and depression. Caregiver Studies with SCI Samples Many findings from the extensive caregiving literature cannot easily be transferred to caregivers of people with SCI for three reasons: 1) Most studies focus on caregivers of people with Alzheimer s disease or dementia Because these caregiver stress/depression findings typically are associated with the level of cognitive function of the care recipient (Miller & Cafasso, 1992; Marchi-Jones, Murphy, & Rousseau, 1996), they may not apply to individuals with SCI, whose care needs are more typically associated with physical limitations than with cognitive deficits. Because caring for a person with SCI is likely to entail a great deal of lifting during transfers to and from bed, wheelchair and car, SCI caregivers may be at risk for incurring injury. Therefore, those studies that 30

PAGE 46

report an association between the physical act of caring and emotional stress or depression, may possibly permit generalizations to SCI caregivers. 2) The mean caregiver age in the majority of studies is 60 years or older. The highest occurrence of SCI is in the 16-30 age group (National Spinal Cord Injury Statistical Center, 2003). In addition, younger caregivers of partners with dementia are more unhappy and resentful of the caregiving role than are older individuals (Fitting, Rabins, Lucas, & Eastham, 1986). Younger spouses providing care to their younger partners with SCI may therefore demonstrate higher levels of depression or stress than reported in the overall caregiver literature. 3) Alzheimer's disease and other pathologies that lead to a need for care typically develop gradually over time, possibly allowing spouse caregivers an opportunity to adjust to changes in function and behaviors of their partners. Furthermore, many of these disease processes generally result in death within one decade of onset (Guttman, Altman, & Nielsen, 1999; Wolfson et al., 2001; Alzheimer's Association, 2003; eMedicine Consumer Health, 2903). Conversely, traumatic spinal cord injury results in an immediate change in functional status and need for assistance. Furthermore, the life expectancy for individuals with SCI has been steadily increasing over the last five decades (Frankel et al., 1998), and many people can expect to live several decades with SCI. Caregivers, therefore, may be 31

PAGE 47

in a position of providing assistance for lengthy periods of time and more likely to experience stress, depression or incur physical injury from performing caregiving tasks over several years. Because SCI survivors and their spouses are younger at the onset of disability and have different caregiving needs than older individuals with disabilities the existing models of caregiving may not apply (Weitzenkamp et al., 1997). Therefore, studying caregiving in SCI provides us with an excellent opportunity to explore the long-term consequences of caregiving. It is unfortunate that there is relatively little research examining caregiving in SCI; and most SCI studies are limited by small sample sizes or combine caregiving spouses with other caregiving family members. However, in spite of such limitations, several interesting findings are reported. In a study of67 primary caregivers (57 spouses and 10 other relatives) ofmiddle aged and elderly individuals with SCI Decker and colleagues (1989) found no statistically significant differences in levels of depression between caregivers and the general population They examined numerous theoretical constructs including perceived health status, perceived control and perceived burden. Caregivers were also asked to assess the burden of caring for the person with disability This study found that individuals with a perception of better overall general health status had less 32

PAGE 48

depression and greater life satisfaction. Greater perceived control over their lives was also associated with higher levels of life satisfaction and less depression. Those who perceived greater burden were more likely to have more depression and diminished life satisfaction. According to Stull et al. (1994), the absence of a comparison group is a problem with caregiver burden research in all of the studies described previously. Family members who are not caregivers do not experience the caregiving burden, which makes comparisons between caregivers and noncaregivers impossible. Studies ofburden require caregivers to assess the amount of time spent on care giving and rate their level ofburden accordingly; based on the assumption that if they spend many hours each day on caregiving, they must be burdened The authors, therefore, suggest that using measures of well-being instead of burden makes it possible to compare caregivers with noncaregivers. A recent study of spouses whose partners have lived many years with SCI attempted to address this gap. Weitzenkamp and colleagues ( 1997) investigated 124 British married couples in which one partner had a spinal cord injury of at least 20 years' duration. Comparisons of caregiving spouses with non-caregiving spouses found overall depression scores as measured by the Center for Epidemiological Studies 33

PAGE 49

Depression Scale (CES-D) significantly higher for the former group than the latter. Stress scores, however were not significantly different, although caregiving spouses reported greater stress than those not providing care. Although there was a strong overlap between the SCI survivors' level (severity) of injury, and their resulting dependency, this factor failed to explain the differences in stress scores between spouses who provided care and those who did not. In addition, the study found that significantly more care giving spouses reported feelings of burnout, fatigue, anger and resentment than did their noncaregiving counterparts. The caregiving spouses also scored significantly higher on measures of depressive affect, and reported higher scores on measures of stress compared to noncaregivers (Weitzenkamp et al., 1997). Since this study sample was British, and no similar studies with U.S. samples have been conducted, it is not possible to generalize these findings to U.S. SCI caregiver samples. Shewchuk and colleagues ( 1998) examined the dynamic processes involved in health and adjustment of family caregivers of people with SCI in the early time period post injury. They assessed 62 caregivers (8 men, 54 women) of SCI survivors during initial rehabilitation and at one, six and 12 months post-discharge. Caregiving experiences were measured using Pearlin's model of caregiving (Pearlin et al., 1990). Depressive behavior was measured with the CES-D. Physical symptoms were 34

PAGE 50

assessed, as were anxiety and positive/negative affect. Lower levels of expressive support-the degree of help and support from friends and relatives-contributed to the development of depressive behavior and anxiety in caregivers over time. Overload, loss of self, and management of meaning were umelated to the development of distress. Based on changes in depression scores on the CES-D between the initial and final assessments, this study found an association between increased depression and both younger patient age and low positive affect of the caregivers. All of these findings suggest increasing depressive symptomatology as time passes, which is counter to clinical expectations according to the authors. Unfortunately, no information is available in this or any of the caregiver studies described above regarding the amount of time spent care giving and the impact that extra caregiving has on stress, depression, or burden. Although Weitzenkamp and colleagues (1997) dichotomized their sample into caregiving and non-caregiving spouses they did not explore differences in outcomes based on the extent of care provided. Studies of caregiving in general, and specifically in SCI, have examined caregiver distress and strain within the context of various stress or burden models. Few other theoretical avenues have been explored particularly those that evaluate the nature of the caregiver-care recipient relationship as a component of caregiver distress. Clearly, 35

PAGE 51

the quality of such relationships may be a meaningful factor in explaining caregiver distress, making exploration of alternative theories a necessary and compelling option. Attachment Theory Because of the interplay of numerous factors affecting caregiver health and distress, designing theoretically-driven research is extremely complex, and it is unlikely that a single theory or model can adequately address all salient issues. However, one theory that may provide the 'glue' to understanding the multidimensional issues related to caregiving both the positive and the negative is attachment theory. Adult attachment theories developed from the idea that attachment is a universal, enduring affective tie between a given child and a specified adult (Bowlby, 1980). More broadly, Bowlby suggested that attachment theory is "a way of conceptualizing the propensity of human beings to make strong affectional bonds to particular others and of explaining the many forms of emotional distress and personality disturbance including anxiety, anger, depression and emotional detachment, to which unwilling separation and loss give rise" (Bowlby, 1977). Expanding this concept of attachment throughout one's life, Bowlby posited that "there is a strong causal relationship between an individual's experiences with his parents and his later capacity to make 36

PAGE 52

affectional bonds." By affectional bonds, Bowlby was referring to particularly close ties "in which the partner is important as a unique individual and interchangeable with none other" (Ainsworth, 1989). Specifically, Bowlby suggested that "the behavior of parents, and of anyone else in a care giving role (emphasis added), is complementary to attachment behavior" (Bowlby, 1977). Insight into how early attachment and subsequent romantic relationships are linked grew out of the work of Mary Ainsworth, a colleague of Bowlby's. She stated that attachments occur at all ages, and these attachments are durable, "even under the impact of adverse conditions" (Ainsworth, 1969). Influenced by Bowlby, Ainsworth expanded the theory in her studies of infant-mother relationships in Uganda and Baltimore. It was from this work that she identified three attachment 'types'-secure, insecure (anxious), and 'not-yet attached' (ambivalent) (Waters & Cummings, 2000). Useful to the examination of spousal caregiving relationships is Ainsworth's assertion that "attachment is a synonym of love, dependency is not" and posed the question, Does a phobic wife who clings to her husband and constantly seeks his proximity love him more than a woman who is less neurotic and more competent loves her husband? The answer is clearly, "No, not necessarily." Indeed, a clinician might infer that the phobic wife loves less than the healthier wife. (Ainsworth, 1969, p. 39). 37

PAGE 53

While much of the work discussed by Bowlby and Ainsworth focuses on infant attachment, others have suggested that attachment theory has relevance across the life span (Waters & Cummings, 2000). Hazan and Shaver (1987) carried out further work regarding adult attachment. They theorized that adult romantic relationships differ from infant-caregiver relationships in some significant ways; romantic love, however, is fundamentally an attachment process through which affectional bonds are formed (Simpson & Rholes, 1997). More work on adult attachment has grown out of a body of research based on Hazan and Shaver's proposed tripartite classification of adult attachment styles in romantic relationships. These three styles, similar to Ainsworth's initial classification of attachment are 1) the secure style, which is typical of healthy relationships in which people feel inclined to be available and reliable, and is characterized by positive attitudes toward intimacy and cognitive flexibility; 2) the anxious-ambivalent style, which is typical of those who express a mixture of anger and disappointment as well as dinginess toward others, or demonstrate rigid or hypervigilant behaviors; and 3) the avoidant style which typifies those who tend to avoid and appear emotionally detached from others, and may emphasize their selfreliance as compensation for their reluctance to rely on others (Bowlby, 1988) (Hazan & Shaver, 1994; Mikulincer & Florian, 1999). Attachment theory researchers in the field of social cognition assume that attachment in adulthood exists primarily within romantic relationships, and differences in the quality of those relationships determine 38

PAGE 54

differences in attachment style (Stein, Jacobs, Ferguson, Allen, & Fonagy, 1998). It has been suggested, however, that being married may itself influence attachment style, and that attachment styles move toward security in more long-term or committed relationships (Kobak & Hazan, 1991 ). Mikulincer & Florian (1999) examined the association between spouses' self-reports of attachment styles and family cohesion among established married couples. Their previous research found that people endorsing questionnaire items that indicated insecure attachment styles (anxious-ambivalent or avoidant) reported more marital conflict and dissatisfaction than those reporting secure attachment. Findings from this study of 93 couples showed that secure attachment seemed to be related to cohesive and flexible family relationships, whereas spouses who endorsed an avoidant style reported relatively low levels of family cohesiveness and flexibility. Those endorsing the anxious-ambivalent style reported high family cohesion but low adaptability. Attachment theory has not yet been investigated in the context of adult caregiving relationships following the onset of disability. Several factors may complicate this type of research. The post-SCI divorce rate may be as high as 48% (Kreuter, Sullivan, Dahllof, & Siosteen, 1998), similar to that of the general population. In addition, 39

PAGE 55

couples counseling, a seemingly important aspect of the initial rehabilitation following injury, may be overshadowed by more immediately critical issues such as home modifications, financing of equipment, and general adjustment to life with a disability. Although it would be interesting to examine attachment style and marital stability-regardless of whether or not caregiving is an aspect of post-injury relationships such a study needs to be conducted longitudinally over many years, and therefore cannot be the focus of this investigation. Examining attachment retrospectively among individuals who have separated or divorced following injury is also difficult, as any assessment of pre-separation attachment is likely to be colored by factors contributing to the marital demise. Studying attachment among couples who remain together for many years following injury provides an important first step. This investigation examines the degree to which attachment style mediates stress, depression and burden for spouses of men with SCI. Similar attachment styles in couples may be only one part of the picture, however. Qualitative Aspects of Caregiving While the studies described in the preceding discussion identified a number of predictors and outcomes related to physical and psychological caregiver distress, few have investigated the more qualitative aspects of caregiving. A qualitative research approach can generate more exploratory (as opposed to confirmatory) information. In 40

PAGE 56

addition, a qualitative approach takes advantage of the rich information provided by those living the experience of marriage and spinal cord injury. It enables us to learn what matters most to these caregivers, not what researchers believe matters. It can also help explore the reasons for particular statistical relationships between variables Clearly, one challenge in assessing these qualitative issues involves recognizing the high degree of subjectivity and difficulty measuring the impact of non-quantitative factors on caregiver distress. Most quantitative measures of caregiver issues focus on factors that are more negative-stress, depression, and perceived burden. Few studies have attempted to identify the positive side of caregiving, and even fewer have incorporated qualitative methodologies to explore the experience of caregiving. Schulz and colleagues (1988) reported that the negative effects of caregiving may be counteracted in part by the benefits of caregiving. In their study, although caregivers reported a great deal of distress associated with caregiving, they also reported a number of beneficial effects. Almost 75% reported that caregiving made them feel useful. Spouses, in particular, often viewed caregiving as contributing to their self worth. Caregiving may also improve relationships between caregivers and care recipients and provide company for caregivers. However, this could be a tenuous situation, if changes in the nature of close and loving relationships put the participants 41

PAGE 57

at risk when one partner is required to assume greater responsibilities in the caregiving role. A qualitative study by Boykin and Winland-Brown (1995) exemplified the many struggles and feelings experienced by caregivers the conflicts between love and frustration. Respondents indicated that the love of their family member was the impetus for the depth of the daily commitment to provide the care necessary moment to moment, but caregivers often struggled with guilt as they tried to be genuinely with and to retain respect and dignity ofloved ones. Although based on only a very small sample this study clearly illustrated both the positive and negative side of caregiving. LeBlanc and colleagues ( 1997) suggested that studies should elaborate how caregiving effects spill over to other domains of people's lives, describing in more depth the varied context in which stressful events and roles result in a proliferation of stress. Zarit and colleagues (1986) found that caregivers learned to manage problems more effectively, or they simply did not let problems bother them any more as time went on. In addition it has been suggested that pre-illness relationships between the caregiver and care recipient may have an impact on the psychological impact of providing care to an ill or disabled family member (Williamson, Shaffer, & Schulz, 1998). Studies to date have paid little attention to the prolonged or cumulative 42

PAGE 58

consequences of being exposed to the stresses of caregiving over an extended period of time. Up to the present, much of the SCI caregiver research has focused solely on quantifiable aspects of stress and depression in caregiving, leaving a considerable gap in our understanding of the caregiving relationship in SCI. It has been suggested that "when an informant explains himself or herself by recounting a personal history, we gain extraordinary access into that person's subjective experience" (Ochberg, 1988, p. 173). This can be accomplished through some form of narrative. Narrative theory postulates that people attempt to organize their experiences in the form of stories that they see as true (Clark & Standard, 1997). Grant and colleagues (1998) indicated that ... there is strong evidence to suggest that family members talk about caregiving not only as an extension of ordinary family relations a11d everyday life, but also in terms of the purposes served by caregiving. The purposes or ends to which caregivers direct their efforts can be used as a powerful device to explain not only what caregivers do but how they evaluate their experiences. (p. 60) One specific form of narrative is the life story, which is a qualitative research method for gathering information on the subjective essence of one person's entire life (Atkinson, 1998). It begins as a recorded interview, is transcribed, and ends up as a flowing narrative, completely in the words of the person telling the story. Through 43

PAGE 59

the life story, information about an individual can be gathered on three levels of personality description: personality traits, adaptations (including attachment styles), and identity (McAdams, 1990). Details regarding the nature of relationships may also emerge in the telling of life stories. With regard to the ways in which stories are told there is "a finite number of basic story forms that people tend to adopt in creating personal myths" (McAdams, 1993, p. 50). These myths involve "an imaginative reconstruction of the past in light of an envisioned future" (McAdams, 1993, p. 53). The common characters present in these personal myths have been described by McAdams (1993) as falling into categories of agency and communion, with agentic characters described as those who "act, think and feel in agentic ways" (p. 134). These individuals are described as those who are "doers" and who seek to control, create, understand or act in other assertive ways. McAdams has classified these individuals as the warrior, the traveler, the sage and the maker. The characters of communion, on the other hand, are those "who act, think and feel in communal ways" (p. 148). Their characteristics are more nurturing, passionate, or altruistic, and this classification includes the lover, the caregiver, the friend and the ritualist. In addition McAdams suggests that there are individuals who are high in both agentic and communal characteristics, and those who are low in both The "imago or character, types described by McAdams are shown in Figure 3. 44

PAGE 60

According to McAdams, imagoes "may personify aspects of who you believe you are now, who you were, who you might be in the future, who you wish you were, or who you fear you might become" (p. 127). In telling one's life story, any one or a combination of imagoes may be evident. Figure 3 Common Characters in Personal Myth ....... : : 'Agency .. warrior ,*'' :fh e .. The.Sage Communion The E over ::.: The : The : Eriend Th d rutualist : '" ., ... ,.f., -':.'. Low iJ!: Agency, Comnl'union Th e The Survivor (''. 'f .: An interesting linkage of the life story interview with attachment theory was explored by McAdams and colleagues (1997), who used Life Story Interviews in a study of 70 adults. The focus of the study was to determine how two groups individuals who 45

PAGE 61

were more generative2 versus those who were less generative-narrated their lives. In this study, the authors identified a key theme of"childhood attachment," based on Ainsworth's work, and assessed the extent to which attachments to mother, father, siblings, grandparents, school friends and church friends were expressed. The findings of these attachments were quantified to categorize each relationship into an attachment style of anxious-ambivalent, avoidant, or secure. Of note was the fact that there were no significant differences between the generative and non-generative groups for childhood attachments suggesting that highly generative adults "do not reconstruct their childhoods as any more secure in key interpersonal relationships than do less generative adults" (p. 687). It was clear from the narratives that a combination of negative and positive factors or "scenes" played out in each ofthe life stories, regardless of attachment or generativity, with positive outcomes often emerging from more negative events. The use of the Life Story Interview in SCI research is not unprecedented. Duggan and Dijkers (1999) used this format to interview 40 individuals with SCI regarding their quality of life. They noted that for most of their study participants the SCI was viewed as a serious disruption in the life story. Key themes that emerged included elation after recovering something felt to have been lost (such as movement in the 2 G.enerativity refers to an adult's concern for and commitment to the well-being of the next generation (McAdams et al., 1997) 46

PAGE 62

hands a previously held job); affirmation that their lives had meaning for themselves and others; accomplishments achieved; depictions of loss or threat of loss; issues related to spirituality and transformations in the way individuals saw themselves The value of nmTative was felt beneficial to both clinicians working with individuals with SCI, and to the individuals themselves, who experienced pleasure and satisfaction in sharing their stories with others. Research supports the concept that marital satisfaction is related to similarity of marital attitudes (Plechaty, 1987), similar values and role expectations (Odell & Quinn 1998), greater congruence of what are believed to be important characteristics in marriage (Ferrell, 1 993), similar perceptions of the spiritual character and significance in marriage (Mahoney et al., 1999) and spousal personality congruence (Richard, Wakefield Jr. & Lewak, 1990). Similarly it could be hypothesized that congruence in life story forms is related to greater marital satisfaction. In tum, this increased satisfaction may mediate caregiver stress, depression and burden. Therefore it is anticipated that the rich nature of this narrative research can pro v ide greater understanding of the caregiving experience for spouses of people with SCI. 47

PAGE 63

Specific Aims The current study examines the multifaceted aspects of SCI caregiving in order to gain greater understanding of this complex role and its relationship with health and psychosocial outcomes. Using a mixed methods design (quantitative and qualitative), this study investigates the extent to which perceived stress, depression, burden, reported attachment style and congruence of life stories characterize the caregiving experience in partners of men with SCI. To narrow the existing gaps in caregiver research in SCI, this study addresses issues that have yet to be explored by pursuing the following specific ai'ms: 1. Examine caregiving issues in couples with SCI of long duration. 2. Compare differences in outcomes of stress, depression and burden between caregiving spouses who provide fewer vs. those who provide more hours of caregiving, and between caregiving spouses and their partners with SCI. 3. Explore the impact of attachment styles on caregiver outcomes. 4. Explore the impact oflife story congruence between spouses on caregiver outcomes. Hypotheses 1. Partners of men with SCI will report greater a) stress; b) depression than their injured partners. 48

PAGE 64

2. Partners of men with SCI who provide more than 10 hours per week3 of assistance will report greater a) stress; b) depression; c) burden than those providing 1 0 hours or less of assistance. 3. Older partners of men with SCI will report less a) stress; b) depression; c) burden than younger partners. 4. In couples who demonstrate greater similarity of attachment styles female partners will report less a) stress; b) depression; c) burden than those in couples with less similar attachment styles. 3 14 hours is a reasonable cut point based on expected outcome data from the Consortium for Spinal Cord Medicine in their Clinical Practice Guidelines (Consortium for Spinal Cord Medicine, 1999). The need for assistance increases with higher neurologic levels of injury; the median amount of assistance required for individuals with low level paraplegia is two hours per day. For this study, a more limited amount of personal assistance (1 0 hours per week) is used, averaging the Functional Independence Measure (FIM5M) (Uniform Data System for Medical Rehabilitation,) expected amount indicated for each neurologic level in the guidelines. 49

PAGE 65

CHAPTER3 STUDY DESIGN This thesis utilized a non-experimental cross-sectional prospective cohort survey design to gather quantitative information about care giving and measures of stress, depression, burden and attachment styles. A qualitative component of the study involved in-depth interviews with caregivers and their spouses with SCI utilizing a phenomenological approach. This allowed the participants to describe the ir lived experiences as married couples in which the male partner has a SCI. Participants The study sample initially included couples in which the male partner had a spinal cord injury for at least 10 years, and who had been together since prior to injury. Excluded were individuals who were not in their current relationship prior to their SCI or who had been injured less than 10 years. A ten year post-injury duration was selected in order to examine the long-term impact of caregiving, as most previous research in caregiving and SCI had focused on much shorter post-injury time periods. 50

PAGE 66

Potential participants were identified fi om the Craig Hospital database. This database contains detailed demographic and medical information on all individuals who have been treated at Craig Hospital for rehabilitation and follow-up since 1976. Craig Hospital is a subacute rehabilitation hospital that focuses its treatment programs on people with spinal cord and brain injuries and conditions. Craig has a broad catchment area for people with SCI that extends to all 50 states and some foreign countries. In the 2002-2003 fiscal year 128 individuals were admitted for initial SCI rehabilitation (consistent with previous years). Nearly half of these individuals were discharged to Colorado residences, while the remaining individuals resided in other states or countries Individuals seen at Craig for rehabilitation come from a variety of educational, vocational, ethnic and socioeconomic backgrounds. Those individuals from Colorado are generally representative of the Colorado population demographic profile. Data regarding the marital status of all men with SCI who were at least 10 years post injury at the time of the study inception were extracted from the hospital database. From this database exploration, 451 men met the criteria of being married at injury or had unknown marital status at injury with a presumed valid address or telephone number. Addresses were presumed valid as long as the twice-yearly hospital newsletter mailings (bulk mail) were not returned due to unknown addresses. In 51

PAGE 67

reality, some individuals did not reside at the addresses stored in the hospital database. This initial listing failed to yield the desired final sample size of 128 (based on the power calculations described below), so the inclusion criteria were expanded to include couples in whom the male partner had a spinal cord injury for at least five years, and who had been together since prior to their injury. This modification was felt appropriate as it would still examine caregiving issues over a longer time postinjury than most previous studies. This extension to 5 years post-injury yielded an additional 82 individuals, for a total potential sample frame of 533. Enrollment All recruitment materials as well as an Informed Consent form were submitted to and approved by the HealthONE Alliance Institutional Review Board (IRB), which oversees research activities at Craig Hospital. On April 20, 2001, the IRB at the University of Colorado at Denver accepted the protocol and approval granted by HealthONE. After identifying potentially eligible individuals from the Craig database, the investigator sent invitation letters to these individuals informing them of the study, asking them to indicate whether or not they met the inclusion criteria (with the same female partner since prior to injury), and whether or not they and their partners were interested in participating. Reponses were requested via an eligibility letter and these were returned in pre-stamped envelopes. No identifYing information 52

PAGE 68

about the respondents appeared on the outside of the envelopes, to protect their identity. The study protocol required agreement to participate from both partners in a couple. In cases where one or both partners declined participation, available database information regarding the person with SCI provided information to determine if there was response bias on the basis of age, time since injury and neurological level of injury. Table 3 shows the results from the initial mailing to the 533 potential eligible participants It should be noted that concerted efforts to contact these individuals were undertaken, including utilization of telephone operator assistance, electronic telephone books, and verification of addresses from medical records. Table 3 Potential Eligible Candidates Participation Status N Agreed to participate 137 Did not meet inclusion criteria 104 Did not wish to particiQ_ate Ill Unable to contact* 178 Deceased 3 Total 533 *wrong telephone numbers; disconnected telephones; no answer at various times of days, evenings and weekends, returned mail with no forwarding address 53

PAGE 69

After ascertaining study eligibility and willingness to participate from both partners, the investigator sent a packet via mail containing a second letter explaining the study methodology, two copies of an informed consent (one for each partner), the sets of questionnaires (color coded with yellow forms to be completed by the partner with SCI and green forms to be completed by the spouse), and a stamped return envelope; The couples were asked to complete the questionnaires independently of one another and not share their responses with one another until after completion, if they chose to do so. Participants were asked to return the forms within three weeks of receiving them. For those who did not return the materials, telephone follow-upoccurred four weeks after the mailing. At that time, potential participants were asked to complete the materials within the next 14 days. For those who preferred, telephone interviews were conducted separately with each member of the couple after they returned their signed informed consent forms. Power Analysis Power and sample size calculations evaluated the ability to detect statistically significant results. For all sample size calculations, the alpha level, the probability of falsely rejecting the null hypothesis, was set at 0 05, while power, the probability of correctly rejecting the null hypothesis, was set at 0.8. At a power of .8 and a significance level of .05, a Y2 standard deviation difference in group means for any 54

PAGE 70

continuous normally distributed outcome variable could be detected for two equal groups (spouse and partner with SCI; younger and older caregivers; wives providing 1 0 or fewer hours per week of assistance and those providing more than 1 0 hours per week of assistance; similar or different attachment style) with 128 total subjects. Thus, it appeared that the study would be sufficiently powered to examine the study hypotheses. Qualitative Sample For the Life Story qualitative portion of this study, a convenience subsample of couples was identified from the larger study sample of 13 7 participants. Although convenience sampling is not the most desirable sampling strategy (Patton, 2002a), it was felt that couples selected with this method would fairly represent the stories of men with SCI and their wives. The convenience subsample was identified as those couples who lived within the metropolitan Denver area. Nineteen men with SCI and their spouses were identified from those who responded to the mail/telephone questionnair.e portion of the study as meeting this criterion. Of the 19 eligible couples, nine agreed to participate in the Life Story Interviews (McAdams, 1995). The ten couples who chose not to participate cited reasons of not having time to engage in the interviews, not wishing to be audiotaped, or at least one of the partners was not interested in participating in the lengthy interviews. 55

PAGE 71

The consenting couples agreed to this segment of the study and signed separate informed consent forms. The Life Story Interviews were conducted with each partner separately and were audiotaped. The venue for these interviews was at the discretion of the participants, and all occurred in their homes. Each life story interview lasted from 90-120 minutes. Couples in this phase ofthe study were paid $25.00 each to thank them for their participation. Instruments (See Appendix A) The instruments used in this investigation are described in detail below. Both the individual with SCI and the partner were asked to complete the Center for Epidemiological Studies Depression scale (CES-D) (Radloff, 1977), the Perceived Stress Scale (PSS) (Cohen, Kamarck & Mermelstein 1983), and the Revised Adult Attachment Scale (RAAS) (Collins & Read, 1990). In addition the partner was asked to complete the Caregiver Burden Inventory (CBI) (Novak & Guest, 1989) and a series of questions designed to collect demographic information (some of this information was already available from the partner with SCI through the Craig Hospital database). Based on a pilot test of the questionnaire packets with ten men with SCI and their spouses, the investigator determined that completion of the questionnaires took approximately 30-45 minutes. Upon completing and returning 56

PAGE 72

the various questionnaires, each member of a couple was sent a $20.00 check to thank them for their participation. Center for Epidemiological Studies Depression Scale CCES-D) (Radloff, 1977) The CES-D is a screening instrument used to measure the current level of depressive symptomatology in general or clinical populations. The 20-item self-report scale asks respondents to indicate the number of days during the past week they have experienced symptoms such as feeling fearful, happy or lonely. This instrument yields four subscale scores-depressive affect, well-being, somatic symptoms and interpersonal affect-as well as a total score. For this study, only the total score was calculated. The CES-D score ranges from 0-60, with a higher score indicating more depressive symptomatology. The CES-D has been widely used within community populations, including the elderly, to assess depressive affect and has been shown to correlate significantly with clinical depression ratings (Radloff, 1977; Weissman et al., 1977; Roberts & Vernon, 1983; Radloff & Teri, 1986). Internal consistency reliability assesses the consistency of results across items within a test, and for this instrument is approximately .85 for the general population. The CES-D demonstrates content, concurrent, construct and discriminant validity (Radloff, 1977; Radloff & Teri, 1986). The investigator selected this instrument due to its broad use in the caregiving literature, and previous use in samples of individuals with SCI and their 57

PAGE 73

spouses (Weitzenkamp et al., 1997; Holicky & Charlifue, 1999). Both the individual with SCI and the spouse responded to the CES-D in order to compare differences in their scores. Perceived Stress Scale (PSS) (Cohen et al., 1983) This 14-item instrument measures the perceived stressfulness of various life situations in community samples Using five response options ranging from "never" to "very often," it asks subjects how often they have had particular thoughts or feelings during the past month. Scores for the PSS can range from 0-56, with higher scores indicating more stress. The PSS is reliable (.85) and correlates with life event scores depressive and physical symptomatology, utilization of health services and social anxiety (Cohen et al. 1983). The investigator selected this instrument due to the above reliability as well as its previous use in samples of individuals with SCI (Weitzenkamp et al., 1997). Both the individuals with SCI and their spouses completed this instrument in order to compare their scores. Caregiver Burden Inventory (CBI) (Novak &. Guest, 1989) The CBI a 24-item multi-dimensional instrument, measures the impact of burden on caregivers Analysis of the 24 items exploring their interrelationships reduced the variables to five underlying factors: time-dependence burden, developmental burden, 58

PAGE 74

physical burden, social burden and emotional burden. Responses to each item range from 0 (not at all descriptive) to 3 (very descriptive). Alpha measures of internal consistency range between .73 and .86 (Novak & Guest, 1989). Although originally developed for caregivers of cognitively impaired individuals, the items appear to be applicable for other types of disabilities. Revised Adult Attachment Scale (RAAS) (Collins & Read, 1990) This 18-item instrument has been used to examine the relationship between adult attachment styles and self-esteem, interpersonal attributions, social self-perception, trust, and romantic love (Collins & Read, 1990). Respondents rate items on, 5-poiht Likert-type scales regarding their feelings about romantic relationships. While the scale allows for a continuous assessment of attachment along three dimensions (trust of others, feelings of anxiety in relationships and degree of comfort with closeness and intimacy), it can also be scored to determine three types of attachment style (Depend....:. the extent to which subjects could trust others and depend on them to be available when needed; Anxiety the extent to which subjects felt anxiety in relationships such as feeling abandoned or not loved; and Close the extent to which subjects were comfortable with closeness and intimacy). The RAAS has good internal consistency (Depend= .75; Anxiety= .72 and Close= .69) and reliability (Depend= .71; Anxiety= .52 and Close= .68) (Collins & Read, 1990. 59

PAGE 75

Personal Support/ Assistance Questionnaire A team of health care and research professionals at Craig Hospital developed this questionnaire to capture specific demographic and caregiving information about individuals providing assistance to people with SCI. The investigator used this same instrument in a previous study of people with SCI and their care giving partners in the United Kingdom (Weitzenkamp et al., 1997). In addition to questions about age, education level and relationship to the person with SCI, questions ask about the time spent providing care and details of the types of assistance provided. The instrument includes a 4-point Likert scale (1 =poor; 2=fair; 3=good; 4=excellent) for which the responder indicates her perception of her own current health and that of her partner's current health. Responders also completed a checklist of various physical and emotional consequences related to having a partner with disability. This included items such as "physical stress or strain," "burnout," "isolation/loneliness," "and "fear for the future." A summation of the items checked on this list of 18 issues provided a "problems" score. A higher score indicated more perceived problems. Finally, respondents had the opportunity to consider if there was anything they would change about their lives since their partner's injury (yes/no), and to elaborate what those changes would be. The dichotomous responses to the first 60

PAGE 76

part ofthe question (no=O, yes=1) were used as predictors (independent variables) in the regression analyses. Content analysis explored emerging themes arising from the second portion of the question that described desired changes. Life Story Interviews A framework for this qualitative aspect of the study is McAdams' life story model of adult identity (McAdams, 1990). According to this model, people living in modern societies begin to organize their lives in narrative terms in late adolescence and young adulthood. People create internalized, evolving, and integrative life stories that serve to reconstruct the past and anticipate the future in such a way as to provide a life with a sense of unity and purpose. Like literary constructions, life stories can be analyzed in terms of plots, settings, scenes, characters, and themes. According to McAdams (1993 ), key components of life stories include nuclear episodes (high points, low points, turning points) (p. 259); imagoes (idealized personifications of the self) (p. 122); and an ideological setting (locates the personal myth within a particular ethical, religious, and epistemological time and place, providing a context for the life story) (p. 81 ). 61

PAGE 77

Analyses Quantitative Analysis Completed survey instruments were coded and entered into a Microsoft Access database. Data were converted from Access files into an SPSS data file to perform all quantitative analyses. Comparisons of key variables evaluated the degree to which there were differences between the 13 7 couples who chose to participate and the 111 who were known to be eligible but did not wish to participate. Comparisons of neurologic status of the men with SCI was done using chi-square tests-one comparing paraplegia versus tetraplegia (2x2), one comparing complete versus incomplete neurologic status (2x2), and a third with individuals categorized on the basis of functional and neurological status (3x4). These groups consisted ofthose with functionally complete tetraplegia (TetraABC), those with functionally complete paraplegia (ParaABC) and all those with functional preservation below the level of the spinal cord injury (All D)4 Comparisons of age and time since injury used independent samples t-tests. 4Tetra ABC' refers to those with functionally complete tetraplegia; 'Para ABC' refers to people with functionally complete paraplegia; and individuals with either functionally incomplete paraplegia or tetraplegia are identified as 'All D', according to the International Standards for Neurological and Functional Classification of Spinal Cord Injury (American Spinal Injury Association, 2002). 62

PAGE 78

Caregiver ages, spouse's time since injury, and CES-D, PSS, and CBI score group means were reported using descriptive statistics. Next, in order to address each of the study hypotheses, it was first necessary to compute the various scale scores, according to each specific instrument's specifications. Hypothesis 1: Partners of men with SCI will report greater a) stress; b) depression than their injured partners. To test this hypothesis, the CES-D and PSS scores of the spouses with SCI and their care giving partners were compared using paired t-tests. Hypothesis 2: Partners of men with SCI who provide more than I 0 hours week of assistance will report greater a) stress; b) depression; c) burden than those providing 1 0 hours or less of assistance. The variable "hours of care" was used to categorize the wives into two groups-those providing 10 or fewer hours per week of assistance and those providing more than I 0 hours per week of assistance. This variable was computed as follows: (hours of care per weekday x 5) + (hours of care per weekend day x 2) The sum of this calculation yielded the total number of hours per week of care. Bivariate regression analyses with hours of care per week as the independent variable were conducted. After this variable was dichotomized (provides Sl 0 hours of care and 63

PAGE 79

provides 1 0+ hours of care), differences in mean scores on the CES-D, PSS and CBI were examined using t-tests. In addition, hours of care was used as a continuous variable in the regression analyses described below. Hypothesis 3: Older partners of men with SCI will report less a) stress; b) depression; c) burden than younger pm1ners. Bivariate regression analyses with age as the independent variable were conducted In addition, differences in mean scores between wives who were less than 48 years old and those who were 48 or older were examined using t-tests. Finally, age as a continuous variable was used in the regression analyses described below. Hypothesis 4: In couples where greater similarity of attachment styles is demonstrated, female partners will report less a) stress; b) depression; c) burden than those in couples with less similar attachment styles. The RAAS was analyzed with chi-square tests to identify differences in attachment styles between spouses. Comparisons of CES-D PSS, and CBI scores based on same or different attachment style were analyzed using t-tests. Finally, linear regression analyses were performed in an attempt to discover what variables were associated with worse scores on the main outcomes of stress 64

PAGE 80

depression and burden. For these analyses, each of the three main outcomes (stress, depression, burden) was used as a concurrent predictor of other outcomes (e.g., PSS score to predict CES-D or CBI score, CES-D score to predict PSS or CBI score, CBI score to predict PSS or CES-D score), along with the independent variables: female spouse's age, time since partner's injury, partner's age, neurologic status of paraplegia or tetraplegia (categorical variable), neurologic status of complete or incomplete (categorical variable), hours of assistance provided (daily average time spent in all caregiving tasks), partner's PSS and CES-D scores, total number of items endorsed on the "physical and emotional consequences" section of the caregiver questionnaire (Question # 16), self-perceived health, perceived health of the partner with SCI, whether or not there was anything the female spouse would have changed, and whether or not the attachment styles were the same or different. Most of the predictor variables were selected due to the evidence in the literature that those variables were related to outcomes of stress, depression or burden. The three outcomes of interest and their predictor datasets are shown in Table 4, indicating the primary study measures which were used as either dependent or independent variables in the specific analyses. All independent variables were first correlated with each outcome to identify bivariate significant correlations. Those significant variables were then entered into a 65

PAGE 81

regression model for each outcome (stress, depression burden) using multiple linear regression. Beta coefficients (standardized regression coefficients) were calculated. Betas further from 0 (either positively or negatively) represent a stronger association between the predictor and the outcome. Positive values of a beta represent a positive relationship between the predictor and the outcome where an increase in the predictor value is associated with an increase in the outcome variable value. Conversely negative values of a beta represent a negative relationship between the predictor and the outcome. Table 4 Description of Linear Regression Analyses Outcome Independent variables Stress (PSS), Female spouse's age, time since injury partner's age Depression (CES-D) neurologic status (paraplegia/tetraplegia), neurologic status (complete / incomplete) hours of assistance provided, total Burden (CBI) number of"physical/emotional consequences" items endorsed self-perceived health, perceived health of partner with SCI, whether or not would have changed the past, same or different attachment style Stress Above plus spouse s and own CES-D score, spouse s PSS (PSS) score CBI total and subscale scores Depression Above plus spouse s and own PSS score, spouse's CES-D (CES-D) score, CBI total and subscale scores. Burden Above, plus spouse s and own CES-D score, spouse's and (CBI) own PSS score. 66

PAGE 82

Qualitative Analysis It should be noted that both internal and external validity checks were incorporated into the qualitative analyses. Internal validity is a measure of the trustworthiness of the findings, however, in qualitative research, validity can be conceptualized as credibility. The credibility of the qualitative data in this study rested on the employment of as much triangulation between the coders as possible (given time, scope and budgetary constraints) regarding the emerging themes and concepts. Responses from the Personal Support/ Assistance Questionnaire to the question "Looking back on your life, and knowing what you do now, is there anything in your life since your partner's injury that you would change?" were analyzed in order to identify any common or recurring themes. Some of these themes were determined a priori but additional emerging themes were also examined. Responses were coded by the investigator and two independent research assistants (analyst triangulation). The coders met to compare their individual findings and develop a set of themes. These were then categorized by the primary investigator, and these categories were confirmed with the two research assistants. 67

PAGE 83

Analysis of the audiotaped Life Story interviews used content analysis. According to Patton (2002b ), "content analysis is used to refer to any qualitative reduction and sense-making effort that takes a volume of qualitative material and attempts to identify core consistencies and meanings." The interviews were transcribed in Microsoft Word and imported into the Atlas.ti software package for coding and analysis. The investigator and two independent research assistants scrutinized the data separately, identifying recurring themes or concepts. At the end of this initial process, the three compared their findings. Discrepancies were discussed and, even if mutual agreement was not reached, those themes or concepts were included in the listing. This listing also included all of the consistent themes shared by the three analysts. The primary investigator then re-coded each interview in Atlas.ti with the identified code system. This process was confirmed by the research assistants. Finally, the new or possible explanatory connections between categories/themes and subcategories were identified using axial coding. In this process, the numerous codes were reassembled into larger categories or themes that emerged from the actual interviews as well as a priori themes identified from prior work by McAdams and colleagues ( 1997). Specific themes of self-evaluation, self-efficacy (Bauer & Bonanno, 2001), agencl, and communion6 (McAdams, 2001) were incorporated into 5 Agency includes concepts of strength, power, expansion, mastery, control, dominance, autonomy, separation and independence. 6 Communion includes concepts such as love, friendship, intimacy, sharing, belonging, affiliation, merger, union and nurturance. 68

PAGE 84

the coding scheme. Each theme could have both a positive and negative component. For example, a theme of"family support" would include phrases relating to the beneficial aspects of having supportive relatives (positive family support), as well as text describing a lack of such support (negative family support). Although the sample for the qualitative interviews was quite small, these themes were finally used to assign individuals into specific imagoes (based on McAdams' conceptualization, described on page 45 of this thesis). In addition, caregiving spouses and their partners with SCI were compared with each other to determine if there were thematic differences in their life stories. Specific similarities or differences with regard to peak events, low points, tuming points, important/influential people, and spiritual/religious belief systems were compared. 69

PAGE 85

CHAPTER4 QUANTITATIVE RESULTS This chapter presents the quantitative analyses of study data and tests each of the previously presented study hypotheses. Based on the findings and supported by the literature, additional analyses were conducted to explore possible relationships between various demographic and other independent variables and the selected study outcomes of stress, depression, and burden. The results of the analyses are described in detail below. Sample Bias Analyses In order to detennine if the sample of individuals who participated in this study were different from those who were eligible but chose not to participate, chi-square tests comparing paraplegia vs. tetraplegia, complete vs. incomplete neurologic status, and neurologic group were performed. Age and time since injury were compared using independent samples t-tests. Results are shown in Table 5. As indicated in the table, there were no significant differences in age, duration of injury, or neurologic status between those who participated and those who did not. 70

PAGE 86

Study participants tended to have paraplegia, incomplete injuries or functionally complete paraplegia. They were also younger and injured for a slightly shorter period of time. Therefore, it appeared that individuals who participated in the study did not differ from those who did not. Table 5 Comparison of Characteristics of Individuals Included and Not Included in the Study Characteristics In Stud Not In Stud 71

PAGE 87

Caregiver Study Sample Characteristics Table 6 presents the demographic characteristics of the wives of men with SCI in the sample. Table 6 Demographic Characteristics of Female Spouses in the Study 1% 44 32% 60 44% 13 10% 19 14% This sample of spouses was relatively well-educated, with most having attended college for some period of time. They were also, on average, approximately 2 years younger than their partners with SCI. Although the mean duration of as shown in Table 5, was nearly 12 years, the mean length oftime providing care was slightly shorter. 72

PAGE 88

Tests of Study Hypotheses Hypothesis 1: Partners of men with SCI will report greater a) stress; b) depression than their injured partners Weitzenkamp and colleagues (1997), in a study of British individuals with SCI and their spouses, found depression to be greater among the spouses than in their injured partners, but there were no significant differences in stress scores. The current study explored these relationships in a different population using the same assessment instruments. Scores on the PSS and CES-D were compared for the men with SCI and their partners using paired samples t-tests. Individuals with any missing data were excluded from each analysis. Table 7 shows the results of the paired t-test analysis. Similar to the British study, there were no significant differences in PSS scores, but there was a statistically significant difference in CES-D scores; however, not in the expected direction. In this sample, the men with SCI had significantly higher scores on the CES-D than their partners It is notable that the current study sample was younger and injured, on average, a shorter time, compared to the British sample, which may account for the contradictory findings on the CES-D. 73

PAGE 89

Table 7 Comparison ofPSS and CES-D Scores Between Men with SCI and their Partners Men with SCI Female Partners Mean s.d. Mean s.d. PSS 21.47 8.15 21.57 8.61 CES-D 15.43 12.16 12.35 9.95 Hypothesis 2: Partners of men with SCI who provide more than 1 0 hours per week of assistance will report greater a) stress; b) depression; c) burden than those providing 1 0 hours or less of assistance N 123 105 Sig. .906 .018 There is evidence suggesting that caregivers who provide more assistance (due to severity of disability or other related factors) may be at greater risk for developing stress, depression or subjective feelings of burden (Zarit et al., 1986; Gaynor, 1990; Miller et al., 1995; Leblanc et al., 1997; Weitzenkamp et al., 1997). To evaluate this, bivariate linear regression analyses of scores on the PSS, CES-D and CBI, with hours of care provided per week as the independent variable, were performed. Results of these analyses are shown in Table 8. Both the CES-D and PSS scores were significantly related to hours of care, but the CBI total score was not. It is notable that the distribution of hours of care was positively skewed, with 24 of the spouses 74

PAGE 90

reporting they provided no hours of care per week. In addition, the CBI total scores were also positively skewed, with 60 spouses reporting no burden. Table 8 Bivariate Regression Analysis of Hours Per Week Caregiving With CES-D, PSS and CBI Scores p Sig. CES-D Score .192 .044 PSS Score .187 .042 CBI Total Score .098 .276 CES-D, PSS and CBI total and individual subscale scores were then compared for the spouses providing an average of 10 or fewer hours per week of assistance to those providing more than 10 hours per week of assistance to their spouses with SCI (the rationale for selecting 10 hours as a cut point is provided on page 50). The analysis was done using an independent samples t-test with hours of care as the factor. Results are shown in Table 9. When dichotomizing hours of care per week, there were no statistically significant differences in either the PSS or CES-D scores between the two groups, however, there was a trend toward greater reported stress and more depressive symptomatology among the spouses providing more than 10 hours of assistance per week to their partners with SCI. The CBI total score and all subscales, with the exception of 75

PAGE 91

"Emotional Burden, were significantly different. Those spouses providing more than 10 hours of assistance to their partners reported greater time, developmental, physical and social burden than those providing 1 0 or fewer hours per week of assistance. There was, however, no significant trend of greater reported emotional burden among spouses providing more assistance. It is possible that the selected cut-point of 10 hours per week influenced these findings. Table 9 Comparison of PSS, CES-D and CBI Scores by Hours Per Week of Assistance Provided to Spouse with SCI <10 Hours/Week >10 Hours/Week Mean Sd N Mean Sd N Sig. PSS 19.8 7.4 58 22.5 9.1 65 .233 CES-D 10.5 9.1 56 13.9 9 9 56 .959 CBI-Time 2.2 2.8 60 4.9 5.0 69 .000 CHI-Development 1.9 3.2 60 3.8 4.6 70 .000 CBIPhysical 1.1 2.2 60 3.4 4.5 69 .000 CBI-Social 1.0 2.1 60 1.9 3 3 69 .001 CHI-Emotional 0.6 1.5 60 0.8 1.5 70 .258 CBI (Total) 6.6 9.9 60 13.9 15.6 67 .000 76

PAGE 92

Hypothesis 3: Older partners of men with SCI will report less a) stress; b) depression; c) burden than younger partners Previous studies have suggested that younger caregivers of people with Alzheimer's disease fare worse in terms of psychosocial outcomes than do older caregivers (Fitting et al., 1986; Zarit et al., 1980). In order to determine if this phenomenon is true in SCI, caregiver age was explored to test the above hypothesis. First, bivariate regressions of stress, depression and burden, with age as the independent variable, were performed. These analyses yielded no statistically significant results, as shown in Table 10. Table 10 Bivariate Regression Analysis of Female Spouse's Age With CES-D, PSS and CBI Scores p Sig CES-D Score .064 .491 PSS Score .086 .338 CBI Total Score .120 .169 The ages of partners were then dichotomized into two groups; those less than or equal to 48 years old and those greater than or equal to 49 years The cutoff age of 48 represented the median age of the spouses. The analysis used the independent samples t-test with caregiver age as the factor. Results of this analysis are shown in Table 11. 77

PAGE 93

Table 11 Comparison ofPSS, CES-D and CBI Scores by Two Age Groups <48 Years Old >49 Years Old Mean Sd N Mean sd N Sig. PSS 20.9 7.7 63 22.3 9.6 64 .019 CES-D 11.7 10.1 60 13.2. 9.8 58 .925 CBI-Time 2.9 4.0 69 4.3 4.5 66 .084 CHI-Development 2.5 3.8 69 3.2 4.3 67 .069 CBIPhysical 1.9 3.3 69 2.6 4.1 66 .076 CBI-Social 1.2 2.3 69 1.7 3.3 66 .017 CBI-EmotionaJ 0.6 1.4 69 0.8 1.7 67 .224 CBI (Total) 9.0 13.2 69 11.7 14. 0 64 .411 In the analysis by two age groups the PSS scores differed significantly, but not in the hypothesized direction, with the older spouses exhibiting greater stress than the younger spouses The only other significant difference was in the CBI Social Burden subscale but again older spouses reported greater burden and this finding was not in the hypothesized direction. There were non-significant trends toward the older spouses exhibiting greater depressive symptomatology and greater CBI total, time, developmental physical and emotional burden scores. 78

PAGE 94

Hypothesis 4: In couples where greater similarity of attachment styles is demonstrated, female partners will report less a) stress; b) depression; c) burden than those in couples with less similar attachment styles Subjects with SCI and their spouses were assigned to the attachment style category in which they scored highest. In the event that the subjects with SCI and their spouses' highest scores were the same for two different categories, those two groups described the attachment style. For example, the three key categories for this instrument are "anxious," "close," and "dependent." In cases with tied high scores, the additional categories were "anxious/close," "anxious/dependent," "and "close/dependent.'' Table 12 indicates the numbers of pairs in each ofthe above categories, as well as the numbers of men with SCI and the numbers ofwives in each category. Data were missing for one of the wives, so the couple was excluded from this analysis. A couple was considered "matched" if a combined style such as "anxious/close" matched either an "anxious" or "close" style in the partner. 79

PAGE 95

Table 12 Attachment Styles of Individuals and Couples Attachment Style N (%)Men with SCI Note: Percentages rounded to the nearest 1% N (%)Wives N (%)Pairs Clearly, the majority of both men and women were categorized as having a close attachment style, followed by dependent styles. Within relationships, the greatest percentage of couples did not have a similar attachment style, but for those with a similar style, most couples were categorized as close. This finding may be explained by the long-term nature of these relationships post-SCI. A dichotomous variable was then created based upon whether or not the subjects with SCI and their spouses' attachment styles matched (O=no, 1 =yes). Using an independent samples t-test with attachment style as the factor, the results are shown in Table 13. 80

PAGE 96

Table 13 Comparison of PSS, CES-D and CBI Scores Among Female Partners by Similarity of Attachment Style Same Different Attachment Stvles Attachment S tvles Mean Sd N Mean Sd N Sig. PSS 22.0 9.0 56 21.3 8.5 71 .746. CES-D 12.7 10.3 54 12.2 9.7 64 .586 CBI-Time 3.5 4.0 59 3.7 4.6 76 .132 CHI-Development 2.8 3.8 59 2.9 4.2 77 .212 CBIPhysical 2.5 3.8 59 2.0 3.6 76 .439 CBI-Social 1.5 2.7 59 1.5 2.9 76 .539 CHI-Emotional 0.9 1.8 59 0.6 1.2 77 .019 CBI (Total) 11.1 13.7 59 9.7 13.6 74 .595 In couples where greater similarity of attachment styles was demonstrated, female partners did not report statistically significantly less stress, depression or burden than those in couples with non-similar attachment styles. The only significant finding was for the emotional burden subscale score of the CBI; however, this was not in the hypothesized direction. There were non-significant, seemingly comiter-intuitive trends toward greater depressive symptomatology, greater stress, more physical burden and more social burden in the female spouses whose attachment style matched that of their spouses with SCI. 81

PAGE 97

Regression Analyses Regression analyses identified combinations of variables that were associated with depression, stress, and burden scores. Variables selected for entry into the regression analyses were initia11y chosen based on findings from the literature suggesting relationships between specific independent variables and the outcome variables. Within each regression analysis a subset of these variables was selected if the bivariate correlation, as shown in Table 14, was significantly correlated with the selected outcome (PSS CES-D, CBI total and subscale scores) at the .10 level. For example, higher CES-D scores were significantly correlated with more hours per week spent in caregiving, more items endorsed as being a problem (summation of problems), poorer health of the partner with SCI, poorer health of the female spouse, a positive response to "would have changed the past," a higher score by the spouse with SCI on the CES-D, a higher PSS score for both the spouse with SCI and the female partner a higher CBI total score and higher CBI subscale scores with the exception of the time dependence burden subscale score. 82

PAGE 98

Table 14 Bivariate Pearson Correlation Coefficients for Selected Independent and Outcome Variables Independent Variables CES-D Score PSS Score CBJ Total Score Time since injury .218 Age of partner with SCI .160 Complete (1 )/Incomplete(O) .153 Hours/week caregiver .192 .187 Sum of problems .350 .401 .463 Health of partner with SCI -.354 -.347 -.159 Health of female partner -.531 -.468 -.171 Would change past 205 .315 CES-D of female partner 1 : '."'k .,;:, w;[;!5'< .759 .245 CES-D of male partner .307 .327 .164 PSS of female partner 759 ., < ( ::.: .282 <," .. PSS of male partner .309 .409 CBI total .245 .282 ;';.:.: _:;__ _"_li; CBI time dependence burden .851 CBI developmental burden .236 .295 .919 CBI physical burden .227 .308 .868 CBI social burden .296 .370 .816 CBI emotional burden .204 .251 .728 Table 15 shows the results of the linear regression analysis for the CES-D. In this model, the female spouse's self-perception of health and her PSS score had significant associations with CES-D scores. A better score of self-perceived health was related to less depressive symptomatology, as evidenced by a lower CES-D score, whereas higher (worse) PSS score was associated with greater depressive symptomatology. 83

PAGE 99

Table 15 Regression Analysis for Female Spouses' CES-D Scores Independent Variables p t Sig. Adj. R2 Total hours per week caregiving -.042 -0.542 .590 Summation of problems .134 1.285 .203 Health of partner with SCI .031 0.333 .740 Health of female partner -.313 -3.511 .001 Would change past -.117 -1.403 .164 Male partner's CES-D score .218 1.552 .125 Female partner's PSS score .605 6.742 .000 .534 Male partner's PSS score -.185 -1.333 .187 CBI total score .590 1.450 .151 CBI Developmental burden -.246 -1.157 .251 CBI Physical burden -.354 -1.776 .080 CBI Social burden -.119 -0.744 .459 CBI Emotional burden .040 0.319 .751 The above regression was also performed using all independent variables, not just those that were significant in the bivariate correlation. This analysis resulted in the same model and adjusted R2 Missing values were then imputed using mean substitution, and regression analysis with variables having significant bivariate correlations was again performed. This analysis yielded a model that included the same independent variables as that without mean substitution. All further regressions for outcome variables were thus performed without mean substitution. 84

PAGE 100

The results ofthe regression analysis for the PSS scores are shown in Table 16. In this model, a higher PSS score for the male partner and a higher score on the CES-D for the female spouse were significantly associated with greater stress. Table 16 Regression Analysis for Female Spouses' PSS Scores Independent Variables p t Sig. Adj. R2 Total hours per week caregiving -.044 -0.561 .576 Summation of problems .053 0.504 .616 Health of partner with SCI -.064 -0.692 .491 Health of female partner .006 0.059 .953 Would change past .136 1.634 .106 Male partner's CES-D-score -.197 -1.397 .166 Female partner's CES-D score .604 6.742 000 .535 Male partner's PSS score .322 2.384 .020 CBI total score -.209 -0.508 .613 CBI Developmental burden .088 0.409 .683 CBI Physical burden .150 0.738 .463 CBI Social burden .049 0.308 .759 CBI Emotional burden -.022 -0.175 .862 For the CBI regression analysis, significantly correlated CBI subscale scores were excluded from the analyses in order to allow other variables to enter into the model. Table 17 shows the results of that analysis. 85

PAGE 101

Table 17 Regression Analysis for CBI Total Scores Independent Variables p t Sig. Adj. R2 Time since injury .172 1.814 .073 Age of partner with SCI .105 1.129 .262 Complete or Incomplete -.028 -0.293 .770 Summation of problems .536 5.318 .000 Health of partner with SCI .090 0.789 .432 .265 Health of female partner -.019 -0.184 .855 Female partner's CES-D score 045 0.324 .747 Male partner s CES-D score 039 0.352 .725 Female partner's PSS score -.018 -0.128 .899 Overall, greater burden appeared to be associated only with a greater number of problems endorsed on the problems listing. Regression analyses were performed for each of the CBI subscale scores to identify demographic variables or variables other than the CES-D or PSS that may be associated with various burden aspects As shown in Table 18, greater time since injury having a partner with tetraplegia (versus paraplegia) and more reported problems were associated with time dependence burden however the adjusted R2 for this model was quite small 86

PAGE 102

suggesting there are other unmeasured factors that contribute to higher scores of time dependence burden. Table 18 Regression Analysis for CBI Time Dependence Burden Independent Variables p t Sig. Adj. R2 Time since injury .225 2.490 .014 Age of partner with SCI .095 1.068 .288 Complete/Incomplete .095 1.012 .313 .128 Para/Tetra .183 2.006 .047 Total hours per week caregiving .100 1.141 .256 Summation of problems .187 2.048 .043 While several variables had significant bivariate correlations with developmental burden, only a greater number of reported problems and longer duration of injury had a significant association, as shown in Table 19. 87

PAGE 103

Table 19 Regression Analysis for CBI Developmental Burden Independent Variables p t Sig. Adj. R2 Summation ofproblems .562 6.585 .000 Time since injury .243 3.010 003 Age of female partner 204 -0 808 .421 Age of partner with SCI .291 1.150 .252 .314 Complete or incomplete .142 1.675 .096 Total hours per week caregiving .004 0 053 .958 Health of partner with SCI .008 0.103 919 Health of female partner .024 0.292 .771 Table 20 shows the results of the regression analysis for the CBI Physical Burden subscale score. More reported problems and poorer caregiver health were significantly associated with greater physical burden with an adjusted R2 of .201 for the model. 88

PAGE 104

Table 20 Regression Analysis for CBI Physical Burden Independent Variables p T Sig. Adj. R2 Time since injury .097 1.145 .255 Complete or incomplete .016 0.183 .855 Summation of problems .351 3.731 .000 .201 Total hours per week caregiving .004 0.045 .965 Health of female partner -.217 -2.477 .015 Would change past .031 0.356 .723 Higher Social Burden subscale scores of the CBI were associated with longer duration of injury and more reported problems, as shown in Table 21. The model for this subscale resulted in an adjusted R2 of .251. Table 21 Regression Analysis for CBI Social Burden Independent Variables p t Sig. Adj. R2 Time since injury .188 2.367 .020 Summation of problems .444 5.015 .000 Total hours per week caregiving .078 0.966 .336 .251 Health of partner with SCI -.036 -0.434 .665 Health of female partner .061 -0.721 .472 Would change past .027 0.328 .743 89

PAGE 105

Finally, the only significantly associated variables with higher CBI Emotional Burden subscale scores were a higher number of reported problems and longer duration of injury, as shown in Table 22. The model for this outcome resulted in an adjusted R2 of .165, suggesting, again, that other unmeasured factors may contribute to emotional burden. Table 22 Regression Analysis for CBI Emotional Burden Independent Variables p t Sig. Adj. R2 Summation of problems .325 3.770 .000 Time since injury .158 1.989 .049 .165 Health of partner with SCI -.059 -0.711 .476 Would change past .121 1.434 .154 The quantitative findings did not consistently support the stated hypotheses. The first hypothesis stated that wives would report more stress and depression than their partners with SCI. The findings indicate that partners of men with SCI did not report greater stress or depression than their injured partners. In fact, the men with SCI had significantly higher depression scores than their spouses. The second hypothesis suggested that more time spent in providing assistance would be related to higher scores on measures of stress, depression and burden. The results indicated that wives who provided more than 1 0 hours per week of assistance did not report greater stress 90

PAGE 106

or depression than those providing 1 0 hours or less of assistance; however, as hypothesized, they did report higher total CBI scores. The third hypothesis stated that older partners of men with SCI would report less stress, depression and burden than younger partners. The results indicate a statistically significant difference between the two groups in the PSS scores, with older wives reporting significantly greater stress than younger wives. In addition, the CBI subscale score of Social Burden was significantly different, again with the older wives reporting greater burden. The final hypothesis was that in couples who demonstrated greater similarity of attachment styles, female partners would report less stress, depression and burden than those in couples with less similar attachment styles. The only significant difference was in the Emotional Burden subscale score of the CBI. However, this finding was counter to that suggested in the hypothesis, with female partners in couples demonstrating more similar attachment styles having higher perceived social burden. Overall, the quantitative findings provide some insight into the factors contributing to outcomes of stress, depression and burden among partners of men with SCI, even though the hypotheses, in general, were not supported. The qualitative aspects of the study, described in the following chapter, help provide a framework within which to further understand and interpret the quantitative findings. 91

PAGE 107

CHAPTER FIVE QUALITATIVE RESULTS Qualitative information collected for this study enabled participants to describe, in their own words, their personal experiences or thoughts related to living with SCI. These data came from two specific sources First, as part ofthe Personal Support/ Assistance Questionnaire, all female partners responded to the question, "Looking back on your life, and knowing what you do now, is there anything in your life since your partner's injury that you would change?" Second, nine couples participated in detailed Life Story Interviews, conducted by a trained research nurse. The interview outline is described in detail in Chapter 3 of this thesis. Personal Support/ Assistance Questionnaire For the analysis of the single response info1mation, three individuals read all comments and assigned a theme "code" to each response. The coders then met and discussed the independently assigned themes. From the 64 responses to this question, the coders demonstrated substantial agreement, with 18 specific themes identified (in some cases, an individual's comments were lengthy and involved more than one theme). In addition, some of the themes did not indicate what the female partner 92

PAGE 108

would have done differently, but represented what she felt was useful or helpful in adapting to the partner's SCI. A listing of the general themes and the number of times a theme appeared is shown in Table 23. Table 23 Themes Regarding What Wives Would Have Changed Theme #of times mentioned Would have encouraged partner to be more independent 13 Would have sought/obtained more help with partner 11 Would have advocated more for self/partner 6 Relied on faith/religion (helpful) 5 Had good support from family, friends (helpful) 5 Would have obtained more education 4 Would have communicated better with partner 4 Would have moved to/built adapted housing sooner 3 Live for the present/enjoy each day (helpful) 3 Would have made more time for self 2 Would have a better sexual relationship with partner after SCI 2 Would have chosen to have children sooner 2 Would have done more activity with partner 2 Would have sought/obtained counseling 2 Would have exhibited more patience with partner 2 Would have sought divorce I Would have sought/obtained necessary equipment sooner 1 Would have chosen not to have children 1 The most prominent themes were a desire to have 1) encouraged more independence in the partner with SCI; and, 2) sought more help in caring for the partner with SCI. Other themes identified areas that contributed to feelings of stress, the need for self93

PAGE 109

advocacy and communication issues In addition, some positive themes were identified, including identifying helpful techniques resources, or having a strong basis in faith. Specifically the two themes that emerged most frequently are described in detail below. Many of the female partners indicated they would have encouraged more independence in their partners with SCI. For example: I would be less helpful and require more from my husband. I feel I have created or permitted much of his dependency. But how can I not do things for him since he is of such good disposition? I would let him be more independent instead of trying to do everything for him. It would have helped him more to know he could do it. I would not have helped as much in the beginning. It caused feelings of resentment on my part. Even ifhe didn't ask for the help I may have done too much. Several of the female partners also indicated that they would have sought or accepted more assistance in caring for their partner as illustrated by the following statements: I would not have taken full responsibility for his care as it was too much for me The stress almost caused me to have a nervous breakdown. I would try not to take on as much responsibility for his care or I would quit work and be his sole care provider. Maintaining a full time job and doing his care has been difficult. Recently we have 94

PAGE 110

an attendant come in on a weekend day to do his bowel program and that has been a great help. I would get help to come in from the beginning. Then he would have been used to it from the start. Now he is used to me and no one else will be as good in his mind. Because ifl had more help I might not be so stressed. We cannot get help, and when we did have help, I still had to help get everything done in the amount of time that the insurance would pay for and, more importantly, to make sure everything was done correctly. Other issues related to increased feelings of stress involved a desire for more counseling or other social support. For example: I would make sure that my husband had counseling. Using the injury as a crutch for not doing things ... using the injury as a continual topic of conversation when he has more positive characteristics to communicate ... I let my husband talk me into moving to Florida because he couldn't take the cold weather in New Jersey any more. I'm not happy here. Everything I had in New Jersey that helped me cope I don't have here-my job, my friends ... There was a relatively consistent theme that addressed the need for self-advocacy being more assertive with regard to health and insurance issues or seeking necessary information, as shown in the following examples: I should have been more aggressive about his shoulder problems that he's complained about since the time of injury. Possibly something could have been done to be in better shape now. 95

PAGE 111

I would have consulted with an attorney sooner to find out what our rights were. We were told things that were not true, which caused a great deal of stress on both of us .I would certainly pursue more avenues for home health care. We were not allowed any help. The first year was almost too much for any spouse to handle. Insurance companies are clueless. Communication, or the lack thereof, was noted by several of the female partners to be an important aspect of their lives. This was mentioned in both negative and positive contexts, as follows: I would have stood my ground sooner and told him sooner of things I could not or would not do. I would also have stated my own need sooner. He's had too much control over me and our son. I would have taken back years of silence and try to be more open about how I felt. The loneliness and isolation may not have lasted as long if we had talked to one another. I would have been more straight-forward and gotten to the point in a confident, loving manner sooner. Once I said 'I'm not going anywhere so we might as well make the most of it' he let go of many defense mechanisms. At that point we both learned how to live and communicate-and with a sense ofhumor most of all! All of the things that made a good marriage before an injury must be amplified after an injury-trust, honesty, communication ... There were numerous positive themes in which the respondents indicated what had been beneficial in helping them adjust to the partner's injury. One consistent theme related to spiritual or religious beliefs. We feel that us doing as good as we do is because of God ... 96

PAGE 112

I will not think we're bad off because after all that we experienced together, we can handle it with the Lord's help. I feel that I was spiritually grounded, which gave me a great deal of acceptance to our new life. Finally, several women indicated the value of spending time with their injured partner, either as something they wished they had done, or something they were actively pursuing. For example: I would go and do more activities with my husband-camping, fishing, outdoor activities. As we get older, it takes more time to bathe, dress, do hygiene of medical equipment. [I would] spend more time doing fun things with each other and our friends and family. Life Story Interviews As described previously, nine couples participated in the Life Story Interviews. A general description of these couples can be seen in Table 24. Compared to the study sample of 128 who did not participate in the Life Story Interviews, analysis of this subgroup shows that the both the male and female partners were slightly younger; the mean duration of injury was approximately 3 years shorter; the number of years spent providing care was approximately 2 years shorter; and mean CES-D and PSS scores were slightly lower. CBI total, and all subscale scores, with the exception of physical 97

PAGE 113

burden, were slightly higher. As a whole, the qualitative sample included individuals with more neurologically severe (Tetra ABC) SCI. None of these differences was statistically significant, suggesting that this qualitative subsample was representative of the larger study sample. Table 24 Description of Qualitative Sample Qualitative Sample Non-Qualitative SaJJ!ple Mean SD Mean SD 46.4 11.3 49.5 11.5 c3 age 47.9 11.4 51.5 11.5 Duration of SCI 8.8 3.8 11.9 5.3 Years providing care 8.1 4.2 10.1 6.8 CES-D 9.9 6.4 12.6 10.1 (5 CES-D 13.5 6.9 15.6 12.5 S! PSS 17.0 7.1 21.9 8.7 (5 PSS 20.8 5.8 21.4 8.3 CBI total 12.6 19,3 10.2 13.2 CBI Time Depend 3.8 4.9 3.6 4.3 CBI Developmental 3.1 4.7 2.8 4.0 CBI Physical 2.1 3.6 2 2 3.7 CBI Social 2.0 4.0 1.4 2.7 CBI Emotional 1.6 3.2 0.6 1.3 (5 Neuro Group All Tetra ABC 33% Para ABC 45% All D 22% (5 Neuro Group Qual Tetra ABC 44% ParaABC44% All D 11% 98

PAGE 114

Analyses of the Life Story Interviews focused on general themes that emerged across all interviews, then on themes arising from the inquiry related to key events (high point, low point, turning point), important people in their lives, plans for the future, stresses in their lives and, finally, on issues related to spiritual beliefs and personal values General Themes: For all participants-both husbands and wives-the most prominent theme throughout the life chapters was having or not having a sense of control over one's situation. The men with SCI noted loss of control and its negative consequences, particularly related to having others provide assistance to them. Many seemed to resent the need for help. The statements of one husband describing his friends' efforts to help him ride an all-terrain vehicle or his trips to visit friends exemplified this situation: When you have two guys tugging at you to get you on the dam thing [ATV] it makes you feel kind of helpless-not helpless, really, but you really don't want the help .. .I've been in some situations like going over to someone's house and you know or you can sense that they don't want to help me get up in the house because there are two steps. For one wife, the loss of her own self and needs became as much of a disability as her husband's SCI, as she stated: 99

PAGE 115

I was starting to be in that wheelchair every bit as much as he was. In eight years, I didn't do anything; I didn't even drive outside by myself at night hardly any more, because there was no place to go. It's really funny how I distanced myself from everything, so my life turned into his life and I was in the wheelchair every bit as much as he was. Another wife, who indicated she had a hard time relinquishing responsibility for her husband's caregiving, echoed this: Being able to go out whenever I want, where I want, how I want, without any guilt. I wish I could ... There's nothing I want more than to have 24 hours all to myself. I can't step out of my role oftaking care of [him]. Even if I had someone to come in and take over for me, I couldn't step out of the house without mentally still being there. Control in the form of strength as a virtue or a liability-was mentioned from different perspectives. One wife indicated she was always the stronger member of the couple, protective of her children, and the one who always was responsible for taking care of things. She indicated it was a role she was tired ofhaving: People tell me all the time how strong I am. I hate it, because I don't want to be strong. I really would love to be one of those incompetent women who needs taking care of, but I'm not. I'd like people to know that I need people. One wife attributed her strength to being a person who could take control of a situation. Although seriously injured in the same accident that paralyzed her husband, she was pivotal in the rescue efforts and described this event and her later sense of 100

PAGE 116

control in making the most of their post-injury relationship in the following statements: After the crash, the responsibility was all on me because [he] was injured so badly. I had to walk for help and at that time it was really hard to walk away because I didn't think I'd see him alive again. That was probably the hardest and most teiTifying thing that's happened in my whole life. That was really difficult. That experience has made me a stronger person. If somebody had said that all this would happen and I'd have to do all this, I would have said, "No, I can't." But it's made me a stronger person. I hope that I appreciate life more than I ever have before, because I don't take a lot for granted. It's made me really strong. Another wife, who, since her husband's injury, continued pursuing a successful career as a businesswoman and a politician, stated: I. .. really am a very strong person. Ifl set my mind to something, nothing can change it. I knew that I was going to have to give up a lot of things and that our life would totally change our maiTied life, our personal life, my life, the way I could do things and couldn't do things and I just knew that everything had to change and it did. But I think being such a strong and independent person has probably been a big plus for me. Another very prominent life story theme was the importance of family, particularly children. Several of the couples indicated that the needs and support of their children were priorities and a source of happiness and strength for the family, as shown in the following statements by both husbands and wives: ... ultimately it's the simple things in life that bring you pleasure-not money and things like that-the simple things like having a child and loving a child. 101

PAGE 117

I was married, had a husband and a child, and it was just the three of us, and that's the same way it was after [his] injury. We just raised our son as best as we could. We did what we needed to just to get through life. I think it's what everyone else does. They do what they can and just go on. We worked really hard to get [our daughter] through school. When her dad got hurt she wanted to quit. She was in her first year of college and she was coming home. We had to absolutely insist she was going back to school to finish her degree. It was a struggle for her, and when she finished it was almost like I had finished college, because no one's ever finished college in our family ... The value of spending time with children after, and perhaps because of, the injury was also noted. For some individuals, family and children were the things that truly mattered. One wife, whose husband had been working a full time job and serving in a volunteer fire department when he was injured, stated: [He] is probably more involved in their lives now than he would have been, because he was always on the go and attended every meeting and every fire call and everything. Now he has gone to school plays and just everything-all the sporting events and he's probably more involved. When he was first injured he was feeling real down about not being able to play with the kids like playing ball and throwing [our daughter] up in the air, because she was still just a baby. Our sister-in-law told him that she had teenagers and that physical time with your kids is a very short time anyways and that it's the emotional time-being able to sit and talk with them that's important. He really feels that closeness with our kids, because that's what they can do. Her husband echoed this statement: I feel I'm really involved in my children's lives. I go to every event that happens choir or music, football, basketball -I go to every 102

PAGE 118

single thing they do. I went to every party they had when they were in elementary school -I know I wouldn't have done that if I hadn't been hurt. I don't think I would have taken the time, just because I was such a workaholic. Clearly, having and raising children was also felt to be an often frustrating, worrisome responsibility that seemed to be noted more by the wives than by their husbands. One family had a son needing surgery, and it was creating concern and stress. In another instance, a mother stated: We're trying to be supportive of him because he's really sad and scared and angry and we're trying to keep him focused on school because of missing school in the fall, and he has to keep his grades up. I think, as I got older, I had an extreme fear of death, probably for myself first and as I got older, it translates now to a fear for my children. There was some desire to protect the children and help them lead a typical family life, by not making them responsible for their fathers' care needs or changing their lives significantly: Very early on I decided it wasn't the kids' responsibility to take care of him, so I really didn't ask them to learn anything because if anything happened to me, I didn't want them to be in the position to have to take over my part. I didn't think it was fair. I knew our boy wouldn't learn any of his care, he's not responsible enough. I knew our daughter would because she's like me. I don't want that kind oflife for her, so I didn't ask them to learn anything. To [my husband] it's very important-and to me, too-to try to have our kids have as nmmal a life as they can and not suffer because of the injury, or not be able to go to college because we can't afford it ... 103

PAGE 119

Certainly, as they got older and saw other kids who could have more active or normal lives, they probably felt it. I think it was probably harder on me to have to literally feed three different mouths, but probably easier for them to adjust because they just grew up with an accessible house and a special van and calling ahead to restaurants to see if it was accessible, and caregivers in and out ofthe house. It's what they're used to ... When they get down about things they may have missed we try and tell them about some of the things that are happening in their lives that wouldn't have been happening had we not had to move out here. They probably would have been at different schools and had different friends and different opportunities, so we try to point those things out to them. Other families, however, incorporated the SCI experience as part of their "normal" routine, having the children take an integral role in providing assistance to their fathers. Two of the men praised the help from their children: My oldest daughter. ... She's always there to help and understand. She was there at Craig, going through some of the classes and things like that. She helps. She's around when I need her. My son ... was always there for me and gave me so much joy. He just really accepted this and never asked why his dad was in a wheelchair. He went with the flow with mom and we did our daily thing. The support of family and friends, or lack thereof, was a recurring theme, and was often associated with better or poorer overall adjustment to the SCI. Positive experiences included having caregiving assistance, outlets for entertainment, and financial or emotional support. The following examples illustrate these supportive 104

PAGE 120

efforts, which ranged from emotional support for the families to instrumental support in the modifications to homes and provision of other needed services: It's been great living next door to my parents. They're a big help. My dad is a self-taught wheelchair mechanic!. .. We have had lots of community support since [his] injury. Lots of things in ourhouse were donated. The work was donated. We only had to pay for materials. The path that goes all the way around the house that makes it accessible for [him] was donated. The landscaping was donatedwork and materials. One woman, whose husband had enjoyed significant neurological recovery in the years following his injury, spoke of the help a friend provided in the early days following the injury: She really helped out when [he] first got hurt. She would bring over. food and stock the freezer here with Hot Pockets and stuff for the kids. She was more like me more like a take-charge person. She got a fundraiser going at work so they could donate money to buy groceries and stuff. They helped buy some ofthe stuffl needed to buy for [him] when he was first hurt. Another wife described a friend who had and continued to be close to her through the injury and through her now troubled relationship with her husband: [She] came every Thursday while we were at Craig and brought us apple bread. I know it sounds like a really dumb thing-apple bread ... Even after [he] came home she would come and she would say, "I know you guys can't go out to dinner but I'll bring you dinner tonight." And she would bring dinner. And the whole time, even to this day, she'll know when something's wrong. 105

PAGE 121

One wife described the support ofher husband's work colleagues: When [he] was injured there was a lot of fear, but I think the company that he worked for at the time he was injured helped us tremendously. We had an excellent support system through them. We didn't have to worry about anything but getting him better and that made a huge difference. We were very lucky with that. The study participants voiced numerous examples of appreciation and gratitude for assistance they received from others. In those instances where support was clearly lacking, both from family or friends, the overall tone of the life story often became much more negative, and stresses appeared to be more apparent. In some cases, there was a feeling of abandonment, while in others, the lack of support was seen as a loss of friendship. In the case of one couple, who, by their own admission, were having troubles in their marriage, the wife spoke of the absence of family support: His mom and sister, they weren't interested. They didn't care, so it was up to me to take care of him, and really, it was my responsibility ... It's funny how they all disappear when you really need them to be there. While we were at Craig, people came to see us and they were all so nice and offered to help. We got home and they just all disappeared, it was hard ... None of the males in our family could even stand to look at him. It was a difficult time you could see the pain on their faces. It was awful that they could just hardly talk to him, they couldn't even look him in the eyes They didn't understand that not only did they abandon him but they abandoned us. One man had been injured in a construction accident, and although he has enjoyed quite a bit of neurological recovery since his injury, he was still unable to participate in activities with his friends that used to bring him enjoyment: 106

PAGE 122

We don't keep in touch any more since the accident. We did a lot together recreationally with hunting and fishing. We did a lot together in that aspect. I miss our relationship. It's weird. I still don't understand why it went down the way it did but I think I've come to terms with it. I'm always expecting to look around the corner and actually see him standing there. I imagine that I see him and don't know what to say. I still haven't come up with an answer to that. Finally, a recurring theme throughout the interviews focused on health issues-primarily those relating to the SCI itself, but also health problems in the wives. Current health concerns were often associated with limitations in the ability to go out in public without fear of having bowel or bladder problems, or were related to functional limitations. Of particular concern on the part of the men with SCI was how their health issues potentially limited their wives' freedom One man indicated that his bowel program had become erratic, limiting his willingness to go out publicly with his wife: I have a lot of stress with my bowel program right now. I hate that. It's a dreaded fear of having an accident. It consumes your whole life. Another man had been experiencing difficulties managing his urinary drainage. Similar to the previous story, he indicated how limiting this was for him and for his wife socially: I was having a problem with catheterization for a while and that was bothering me ... That used to bother me a lot-more than anything. That holds me back from going places during the day because I have to cath four times a day. I have to know where I am going to be when it is time ... I'd have to be here at a certain time, and I know I held back 107

PAGE 123

my wife from doing things and that bothered me. I mean, I let on that I was mad about the cathing, but it was more that I felt like I was holding her back from doing stuff. Health issues were also associated with feelings of fear or worry that the partner with SCI would no longer be able to maintain the same level of independence, participate in enjoyable activities or receive adequate assistance when needed. One man, who had tetraplegia and required help from his wife for many of his self-care activities, stated: Probably the biggest thing that worries me is what I would do if [my wife] couldn't take care of me. That's my biggest fear. What if she died. What would happen? I've told a number of people to not count on me being around A wife, who was the oldest participant in the qualitative study, had concerns about her own health compromising the care of her spouse, who was experiencing numerous health problems of his own. Although he had become able to walk, recent health problems had caused him to again use a wheelchair. I had a knee replacement because I was in a lot of pain and I have a lot of back pain, too and I think a lot of it contributes to when he was in a wheelchair-that s when it all started. Later it was just arthritis added to that. .. things will happen when [he] deteriorates, and I could too, for that matter. I'm 76 years old and all kinds of things could happen. 108

PAGE 124

Critical Events Participants were asked to identify three specific key events in their lives; a peak experience, a nadir experience (low point), and a turning point. For each of these key events, they were asked to indicate what happened, where and with whom they were, what they did and were thinking and feeling in the event, what impact this event had in their life story and what they perceived this event says about them as a person. Peak Experience. Nearly all of the participants could identify a particular time or event that they considered a significant peak. These experiences frequently related to family events such as getting married, the birth of siblings, children or grandchildren, or the successes enjoyed by their children. Two women described the birth oftheir grandchild and child as peak experiences in detail: I think when my grandson was born. My ... [husband] and I were there to see him born. I think that was the happiest I've ever been. I was thinking what a miracle it was. I mean truly-even though I have two children of my own, I don't think you realize what a miracle it is ... It's given me something to look forward to ... It makes me want to keep working, do this or go there it just kind of gives you a new direction in your life. I would have to say the birth of my first child. I think especially in that case because I had wanted a child for so long. Initially, when [he] and I were married, and then he had the accident .. .I thought we might not be able to have any. Then we found out that we could, but then we were dealing with all the effects of the accident. And [he] honestly wasn't ready to have children yet accident or no accident. So I sort of patiently waited and then he finally agreed to try and have kids. I got pregnant and then had a miscarriage ... then we ended up doing in 109

PAGE 125

vitro to get pregnant and I did get pregnant and all went well. With the birth of my first child, it was such a long road to get there, so when it finally happened it was such a huge relief and so unbelievable and wonderful. That would be a peak for me. Another wife described the joy of her wedding and it's meaning to her: The most fun was probably my wedding day. That was because I knew that I had found the person I would spend the rest of my life with. There was no better way to do it. My entire family was with me and that was very important ... Everyone that meant something to me was there. That's why it was the most important day. The feeling was happiness. It was just knowing that everything we were doing at that point in time was right. It was the best decision you had made in your life at the time. One couple had shared sentiments about their peak experiences. For the husband, he indicated that his marriage and birth of his children were peak experiences "which I felt were very enjoyable. I think I felt very-a lot of joy in my life and a lot of pride, I guess, because I felt we had a good family." His wife went into more detail about these experiences, as follows: It was probably when I malTied [my husband]. It was certainly a happy occasion and I had great hopes that we'd have a good marriage, which we have. Of course the birth of our two children it was very special and happy and exciting. They're great kids, too__: they turned out very well. They're well-adjusted people, which makes us feel very proud. Those are the things that made me happy, and then the grandbabies came along and you can't excel that because that's very special ... 110

PAGE 126

Only one of the men indicated the peak experience was getting some degree of neurologic recovery following his injury. He described this event as follows: ... when I first started moving my feet, especially after I didn't think I was going to walk again since the doctors gave me zero chance. I used to lie in my hospital bed and try to move my big toe every night. Then one night I was like "I cannot believe this!" There was no one there to see it move. It was 10:00 at night and everyone was home. I had to wait until the next day to show everyone that I could move my big toe. I could barely move it but I was moving it. At that point I knew I was over the hump. Other peak experiences dealt with accomplishments and independence, expressed by both men and women. One of the husbands had been married prior to his current wife and recalled the time between his marriages as a peak experience: It sounds pretty crazy, but as you can probably guess, I've been married most of my life and probably the two best years I have had were not being married being single. The most fun and peak experience and biggest adventures and being wild and crazy were the two years that I was single ... What that says about me, I would probably say is, I still don t know what I want to be when I grow up. I can still be on the wild side when my chair lets me. For the wives, peak experiences also related to their independence or achievements. One woman described the onset of her independence at an early age: I would say probably moving out of my parent's house was the first thing. I moved out the day I graduated, when I was 1 7 and moved in with some girls. That was probably one of the best times .. .I loved it. I never ever went back ... I made it on my own. I never asked my parents for money ... 111

PAGE 127

Another woman described her peak occurring at a time when her career was blossoming, but her father, to whom she was extremely close, was dying: I would say last year when I got elected as an elected official. I really wanted that. .. in May we found out [my dad] had colon cancer-he said to me that he would be alive to see me win the election and to see me sworn in ... That was really exciting to finally win the election and know that my dad had been there for that. I felt very proud. Nadir Experience. Low points were described as events that engendered feelings of sadness or anguish, often precipitated by personal loss. It was not surprising to find that the injury itself was considered a low point, as it was an event that was sudden, unexpected, and had significant functional consequences. Several of the men recounted specific low experiences related to the injury, most regarding the very early days and weeks following the accident. These episodes incorporated a tone of fear, uncertainty and despair, as described by one man whose injury resulted in tetraplegia, limiting the types of activities he used to enjoy: After I got hurt. I would say at least the first two years after I got hurt. Why didn't I just get hurt so bad that I died? I'd sooner be dead as be in the chair .. .I love hunting and fishing. We used to have horses. I loved backpacking on horses and would just go up on trips. I really enjoyed the outdoors, which I can't do now. Another husband, who was injured while diving in a backyard pool, said, "the low point was breaking my neck. I knew I was paralyzed the minute I hit. I didn't know I was paralyzed permanently." 112

PAGE 128

One man had looked forward to a relaxed retirement, but at the age of 68 was injured in a fall from a ladder: Well, when I had my accident I think I was very low, because at that time I had been retired for about three years and I had planned, pretty much, a longer, healthier retirement. I enjoyed playing golf and that cut out my golf and, as I said before, it restricted our traveling quite a bit. That was a very low point. I was -I don't think I was depressed but I was very discouraged and disappointed. In other cases, the low points centered on the longer term consequences of SCI, but not on the actual events surrounding the initial injury. These examples identified feelings of :frustration or loss. One man indicated: I guess I'm afraid sometimes. It goes back to when I was hurt. I'd b,e scared to be in public sometimes. Not so much people staring at me with me being in a chair or anything like that. But just knowing that I wasn't able to accomplish what I wanted to do at that time-whatever it was. Just to go up the stairs or something like that to take that first step. Especially if I had to go to the restroom or something like that. Another husband was having numerous difficulties interacting with others, including his wife, and indicated the strain on the marriage was increasing. He lamented the inability to have physical contact with his grandchild, and was unable to finish his statement as it was emotionally wrenching for him: One of the low points for me is not being able to interact with other people, and it happens quite often. I'm not sure how to say it, but the intimacy or even just being acknowledged by other people ... And even just intimacy with my wife. Because even when you're with the most intimate person, it's still really hard to get close to somebody, because 113

PAGE 129

you can not get physically close to them. It's hard to touch somebody because you have to reach over and get into position while you're watching out that you don't run over their feet or run into the furniture. That's probably one of the lowest points. Another one is not being able to hold my grandson. That's something you live for. I think that's a low part. I can't do things, you know that you are missing out on things that other men are getting to do, that is probably the most ... For a few of the wives, the injury and consequences of SCI were cited as significant low points, which created feelings of fear, frustration and anger. One woman, whose marital troubles had existed before the SCI, stated: I think when [he] got hurt. I think I was pretty negative about most things. It takes a long time to get over the initial bad feelings, and I don't think you ever really get over it. You just kind of deal with it because you have to, but that doesn't make you feel any less resentful about the way things happened or how it changed things. Another wife, whose husband's injury left him with a very high level of dependency, described her devastation at the news of his injury: Probably the night that [he] was injured, just hearing what the diagnosis was. It was a roller coaster of glad that he was ok as far as being still alive, because I thought he was probably going to die, and fear of what the future was going to be like. Where were we going to end up, somewhere between where we were that night and leading some kind of normal life again. It was probably fear and sadness and a combination of a bunch of other emotions. Several participants voiced other low points to include the loss of loved ones-parents or siblings. These helped exemplifY the fact that many of these individuals 114

PAGE 130

had encountered adversity before the SCI that had a significant impact on their lives. One woman described the loss of her brother: He died the day before he would have turned 17. I was eight years younger than he was. My father wasn't home-he was working, so my brother was the caregiver and the disciplinarian at times. He was just always there. I think maybe because of memories -I was young when he did die-I only remember the good parts. Another woman, who was raised by her widowed mother, recounted how the death of her father affected her, and how she continued to anticipate that good things could be temporary: ... as a child when you lose your father I think it results in giving you certain aspects of your personality that you'll have for the rest of your life. It's not good and it's not bad. I think for me personally there has always kind of been a feeling-because I was 8 when he died and as ,an 8-year-old you're just sort of rolling along in your life and you certainly aren t thinking very deeply. All of a sudden you lose your father and your world is turned upside down and you can't make sense of things ... I think for me, when things are good in my life, I have a tendency to wait for the other shoe to drop because that's what I think I learned as a child. You know, don't get too happy here. Don't get too secure, as if being prepared for something would make it less awful. Perhaps the most poignant low point was identified by one of the wives, who had been caring for her husband for several years and suppressing her own needs. Her story illustrated the despair she ultimately recognized: I woke up one morning and could understand how people could kill themselves. Not how they could, but I understood why people would want to kill themselves. It wasn't even killing themselves-that's the wrong thing to say. It's just one of those bad feelings, not that I wanted to kill myself or wanted anybody to kill me. But maybe I 115

PAGE 131

would just feel better if it all ended. That was the absolute lowest I ever felt.. .I actuaily let my self-confidence get totally out of control. I mean really. At that point I had no self-esteem, no self-confidence. You know, I'm actually a quite confident person, open and big mouthed, but I had turned into this meek little, mindless, non-thinking woman, it was just totally depressing. My whole life and whole self was lost. It was gone. Turning Points. Participants were asked to recall events in their lives that affected the way they saw themselves or changed how they understood themselves. For some, the turning points related to either high or low points and represented some form of growth or personal transformation. A frequently cited turning point related to the accident, as noted by one man, who said: It was when I had my accident. I changed the way I felt I was and things that I could do and things I had planned and I guess I changed to someone who's a lot more reserved or restricted in my activity ... I guess it changed my outlook on what my future should be would be. I tried to do some of the things I had done before and I found out that a lot of those things weren't possible ... His wife commented on the same turning point: I'm sure it was when [he] was injured. I felt that this is why I'm hereto take care of him and I had to deal with it. Because everything up to that point...itjust seemed like everything was rolling along on an even keel and I dido 't have to worry about who I was. I was a mother and a wife and that was great. Then, when this hit I realized that I had to have a new role and would have to do it, you know, the best I could. Only one other wife cited the injury as a turning point, stating that the SCI instigated a significant change in their lives: 116

PAGE 132

The accident that definitely is the turning point. When you go through a tragedy like that something that is so much of a trauma, it would change people most of the time. It changed us. I can tell you that. We prayed and prayed for [him] to be healed and I think God healed us not how we wanted to be healed. We wanted [him] to walk but I think God has done a miracle for both of us from the inside out. For one husband, the SCI completely altered his life. His recounting of his low point, which was the injury, and his inability to identify a clear peak experience were echoed in his sentiments about his turning point: I'm two different people, kind of. There's [me], without the spinal cord injury guy, then there's [me], the spinal cord injured guy .. I've always been kind of out-going, vain, somewhat arrogant. .. After the accident I thought it would still be the same but that didn't prove true. It's kind of hard to act as goofy as you used to because people look you and think, "what's the matter with him? Is he a head trauma or spinal cord injured?" That's the way I perceive it anyways. I probably was the same when I was able bodied. If I saw somebody in a wheelchair having a good time, it was like "is that person off or what?" I didn't know any better. So it's almost as ifi have a double personality. For another man, he felt the injury improved him and made him a better, more tolerant person: .. .I have more patience now than I ever had. I never had any patience before. When I worked and was a supervisor, I expected everybody to work as hard as I worked and do what I did. Now I know that everyone's different. I can take my time and blow things off. Not that I do, but I can ifi want. I realize that the world's not going to end tomorrow ifi don't get the grass cut, you know, or the dishes aren't done. I used to have to have everything done. 117

PAGE 133

Another frequent turning point theme of transformation related to personal growth maturity, achievement or independence unrelated to the SCI. This was a more prominent theme among the wives. One woman had grown up in a rural community with an alcoholic father and expressed her need for independence and growth: ... a turning point for me I would probably say leaving Nebraska and moving out here, because I just left it all. One day I just got fed up with Nebraska ... I came out here and I think that s probably when my life turned around. Because I think I was just going nowhere there. With alcoholism in your family my dad was a very bad alcoholic and with mental illness in your family I think that moving out here turned my life around. I just kind of left it all behind. One wife had been married prior to her current relationship and reported how she recognized significant self-growth after leaving her first husband: I think after my divorce and I had to be on my own with two children. I figured out I could do things for myself. For the longest time I stayed in a marriage that I probably shouldn't have because I was just afraid that I wouldn't be able to find a job and it's just easier not to worry about it or think about it than it is just to get out there and be on your own. Once you're on your own you actually really like it and it isn't nearly as awful as you thought it would be. If anything, it makes you think if you really want to get married again because I can do these things for myself. I don't need someone to do them for me. I think that was the big turning point findipg out that you re not some little weak person who can t stand up for herselfand can't make it to work every day. 118

PAGE 134

Another wife, by her own report, had been fairly wild as a young woman in college, left school to work, and then realized that in order to get a meaningful job, she would need to explore further education: I think that was when I went back to college the second time and I definitely realized certain things about myself at that time. I realized I was ambitious. I did want to be successful in life at certain things ... So college was the way to do that. I sort of fumbled around for a long time before I understood that college was the place for me, so I was very happy to have gotten to that place and have gotten myself together at that point. I was probably 21 when I was back in college, serious about it, and knew that I was on the right path in a professional and career way. In summary, the key points in the lives of these individuals were not, on the whole, necessarily related to the SCI. The highs tended to revolve around family events, while the lows generally represented loss -of loved ones, of physical abilities, or of self-esteem. Turning points covered a variety of events, however, with the exception of those specifically linked to the SCI, most of these episodes occurred at a relatively early point in the participants' lives. Clearly, turning points often related to establishing independence. For the women, the ability to leave home, become meaningfully employed, and enjoy their own activities were important. For the men, physical as well as economic independence were valued. Factors that helped individuals as they changed their lives were perseverance and determination, similar to findings from King and colleagues (2003) in their study of turning points in people with chronic disabilities. In addition, these turning points typically helped the 119

PAGE 135

participants recognize some positive characteristic in themselves, or assisted them in gaining a better understanding of themselves through positive experiences as well as through loss. Significant People Tedeschi and Calhoun (1995a) indicated that "the quality of the relationships with family and friends before and after [a] traumatic event plays a role in determining the likelihood for growth: If good relationships can be maintained or improved, growth may be possible" (p. 94). Approximately two-thirds of this sample reported that parents or grandparents played a significant positive role in their lives. The parental contribution came in the form of instilling values, setting an example, or being generally supportive in the lives of men with SCI and their wives. One woman, whose husband's injury required a great deal of personal assistance as well as a move to a fully accessible home stated: ... my parents. I've been close to them and being an only child, you know, we were always close. After [my husband's accident I really turned to them for support and help with the kids and help with [him]. .. Another wife spoke of her mother with great respect: ... my mother. .. raised four kids as a single parent and did a fantastic job. She's my hero. I admire her very much. She's a great person. 120

PAGE 136

She's very, very honest, very positive about life and I just think she had a hard road in life and handled it really well. That's definitely a significant person in my life-probably the most significant. One woman described her father who she had come to love and respect as the years passed: My dad, just because I don't think he grew up until he was about 40 or 50, but he's someone .. .I strive to be. I'm very proud of who he's become. He takes care of all the neighbors now. He sweeps their yards and he's there for everyone. He's who I strive to bemy hero. One of the men with SCI described how his relationship with his father was not close because his father was "a mean person." For this man, his grandfather stepped into a parental role, and taught him values that he retained: My grandfather became another important figure in my life ... he was more like a father to me. He instilled in me those things that my father was not. I got a lot of discipline from my father, but I learned that life could be fun from my grandfather. I also learned how to treat women from my grandfather, because my father didn't treat women with the respect that my grandfather did. Nearly everyone indicated that they had friends who had made a significant impact on their lives, most notably in providing support and companionship, particularly during difficult times. The men with SCI highly valued the camaraderie shared with their friends, and the continued relationship following the injury, as described in the following statements by three of the study participants: 121

PAGE 137

One of the people that made a big impact in my life is my friend, Carl. I've known Carl since ... He's almost kind of like a big brother to me. He taught me a lot of stuff. I mean, I grew up and didn't have a father. .. I think he kind of helped define me a little bit more, just by watching him. There's just something about him. I like a lot of his personal traits, they helped me through the anger after my injury. Back then I was very aggressive. He's changed my life. He's taught me a lot of things. He's a true friend, even though we only see each other a couple times a year. He still thinks of me and I still think ofhim, and it's kind of nice. I would probably say one of my true friends is Jim. He's pretty much stuck by me ever since I got hurt. He's been a true friend and has always tried to help when he could. He still doesn't understand everything about my injury, but he's more understanding of my position as far as my limitations. I had a close friend who I went through high school together with. We both went into the Army at the same time. We were overseas together and then we split up. We both got out about the same time and we'v_e been very close ever since. I see him at least once a week and we've been very close ... He's probably my closest friend. We've gone through a lot of things. He's had some health problems and I've had my health problems and we've supported each other through those. We're still very good friends. Two couples acknowledged one another as having a significant impact on their lives. For the wives, they identified admiration and respect for their spouses as husbands, parents, and role models. One couple echoed this mutual respect in the following statements: I feel very lucky that we found each other. I think he's just an awesome person. He's the funniest, smartest person that I've ever known ... He's my partner. We're raising a family together. We're going through life together, making decisions together. We're friends. 122

PAGE 138

We talk about things-not only our personal lives but the world. So I guess that's whyhe's significant-because we share life together. The husband's statement, although more modest, reflected his satisfaction with his partner: My wife, obviously, is important. We're around each other quite a bit. We live together-I guess all the standard reasons. I love her. We do most things together. We have a normal marriage. For another couple, their feelings of love and respect for one another were highly evident in their statements: [my husband is] my hero. He's probably one of the strongest people I've ever met in my life ... He's so determined in everything he does and the decisions he makes. We're total opposites ... and I think that's why we work so well together because I bring out his emotion and he fixes my procrastination. He's a hero . my wife-especially what she went through with me with this injury. I know it was hard on her the first few months ... at first she was pretty much the caregiver. She was doing my bowel program and the cathing ... She was coming to the hospital every night. She'd get off work, pick up my mom and she'd be down there every night, for the six or seven weeks that I was down there. I know that had to be rough, you know, to constantly be on the go To get up, go to work, and then spend every night down there until 1 0:00, then come home and go to bed ... We're best friends. We take care of each other. Of course, we fight and have arguments, but that's normal. Other husbands identified characteristics of strength, devotion, and compassion in their wives: 123

PAGE 139

My wife is one of the people that has made a significant difference in my life. We're fortunate that we have a lot of likes that are the samelikes and dislikes that are the same, so we re pretty compatible. She's always been there for me, regardle ss of what I've done either good or bad. Since my injury she's been there all the time for me and it's amazing how much she's meant in my life because I know a few people that wentthrough what I did and their wives couldn t handle it. They took off. My wife hasn t been that way. She's been right by my side and been my support and generally everything we do, we do together. Clearly, for both men and women not only were family members important, but the presence of a good friend who could understand their individual needs also stood out as a dominant feature in their overall satisfaction with their lives. These friends and other supports were generally construed as coping resources for both the men and their wives, but particularly for the wives, who found the outlets provided by friends and other relatives helped relieve some of the stresses of caring for their husbands with SCI. Future Script When asked to consider plans or dreams for the future, family needs again represented an important theme, particularly in those couples with children still living at home. The concept of generativityhaving a concern for and commitment to the well-being of the next generation became apparent in this segment of the life stories. 124

PAGE 140

One wife had three young children who, although often frustrating, provided her with a sense of happiness and fulfillment. Her goals were clearly oriented toward her family, at least for the immediate future: My plan is to be a stay at home mom for the next three to five years, raise my kids in the most positive, happy way that I can, and after that I can go back to work. Self-interests were occasionally mentioned in thinking of the future, but they tended to contain a component of caring for or giving to others. One man had been discouraged by the limitations his injury had placed on previously enjoyed activities, but now had been exposed to a meaningful organization. He stated: I would say looking at the future I'd really like to take this Wheelin' Sportsman as far as it can go. I'm really getting to where I enjoy it and I know when I go to a place, whether it's accessible or not. I know that I am helping other people so that they get to go hunting and it will be accessible to them and hopefully they'll have an ok time. I'd really like to see how far this thing can take me. Another husband indicated he would like to pursue avenues that would help others with disabilities: I thought about working with kids with handicaps. There is a school over on 6th for kids with disabilities. I could stay where I am doing that. Where I work now; they have some classes I could take to learn how to be what they call a group leader. 125

PAGE 141

Maintaining the status quo, enjoying financial security and maintaining health (with occasional discussions of "cure") were also prominent themes in the future scripts. Two of the men indicated that they were fairly satisfied with their lives at present, and really didn't make big plans for the future: I live my life day to day. The plan is in the next 15 years to get enough money in the bank and enough equity so I don't have to worry when I lose my ... benefits ... When I'm 63 I lose the annuity that I have. Basically all I do with that is save ... I do plan. At first I didn't think I was going to live until I was 65 and be 3 7 years in a wheelchair how could you live that long? But now I've met people who are that old in wheelchairs and are hurt basically about the same way I am, and they're not in as good of condition as I am. I take every day one day at a time. To me, I have everything I want. I don't need anything. I'm not saying that I'm extraordinarily wealthy or anything, but if I want $Omething I can go get it ... I'd like to get 100% healed from the injury, and I know that's probably not feasible, but maybe someday it will be. I've never given up hope of not getting everything back. For one of the wives, there was a desire to see her husband recover more function: My dream would be that [he] would be able to walk. That would be a great blessing in our life. If that's not God's plan I guess it won't ever happen. If we're talking about dreams, that's something that I pray about all the time. I really just pray that [he] stays healthy and does as well as he's done. He's usually not sick and doesn't have to go back to the doctor and I just hope we can keep that going. In general, the future scripts were quite modest in tone, with the desire for health and happiness predominating these discussions. 126

PAGE 142

Stresses and Problems Lack of independence was clearly a stressful situation for those husbands who required physical assistance from their wives and others, or who felt that their disability was disruptive. These sentiments were expressed by the men in terms of disempowerment or burden: I guess, and it's my own fault-it's the dependence, the fact that I'm always asking my children and [my wife] to do stuff. It seems like no matter what-that's all I'm doing is asking for something. A lot of times I feel like I'm a burden. I think that's a stress. I think about it a lot, but in the riext breath I'm asking for something. I know they don't mind but, still. .. Probably the biggest thing that worries me is what I would do if [my wife] couldn't take care of me. That's my biggest fear. What if she died? What would happen? I've told a number of people to not com;:tt on me being around. The only one I haven't told is my son because he doesn't believe in suicide. I'm sure I would adjust but I'd have to have strangers coming in. I'd have to become much more independent in ways that probably would be harder on me simply because when I was first hurt I took care of myself a lot more. With back pain now and limitation in movement it's made it a little tougher. I worry about what would happen ifl lost her. The wives, on a similar note, described the stresses of providing assistance, and not having time or energy to satisfY their own needs; or not wishing to relinquish control over the situation, as illustrated in the following comments: Well; one [stress] is being able to have the ability to say, "the hell with it" and get dressed up and go out. Being able to go out whenever I want, where I want, how I want, without any guilt ... that's a real conflict for me and it's real hard for me to, you know, people will ask 127

PAGE 143

me to go out with them after I put [him] to bed but I can t do that. I just can't do it. I wish I could but I just can't do it. The biggest stress always is taking care of [him] beca.use whatever I do, I like to do 100%. I'm not somebody that can just kind of do something halfway I take this seriously and he has to be clean and his clothes have to be clean and he has to be taken care of properly. That's a big stress for me and part of the stress is that I put it on myself. Health issues also constituted an area of stress, particularly for the men with SCI, who indicated that their general outlook was highly dependent on their physical wellbeing. One husband had been experiencing numerous health problems in recent years, stating: Well, my condition-my physical condition__:_ is causing me quite a bit of stress and conflict. The thing is, with what I'm going through right now, I have something that happens, like when I broke my leg or when I had to go through prostate cancer treatments that s really a downer and I feel very discouraged ... I think everything kind of is involved with my physical condition. When things are going good, I feel good, but when things are going bad, it's kind of a downer. I try to keep my spirits up but it s hard. Another man, who no longer used a wheelchair, said: Walking up stairs is a problem for me every day That s a tough one. Cold weather is a problem because I stiffen up and don't know why ... I don't want any more medicine. I'm tired of medicine. While most of these marriages appeared to be quite stable, not all the relationships were solid creating concern and stress. One husband had great concerns that his marriage was failing, and communication between him and his wife was difficult: 128

PAGE 144

... my marriage ... I don't know where it's going to go. I guess since there's so much lack of communication and [my wife has] always been really too tough to talk to without getting all bent out of shape, starting hollering and screaming. I wasn't raised that way and I don't do that. Sure, everybody gets mad and hollers a little bit, but I just feel that if you can't sit down and talk civil to each other, I'd just as soon forget it and go somewhere else. I guess it's tough for me because we've been married 20 years and I still feel like I'm very independent and I can probably still survive and do the stuff that I can do because, right now, [she] doesn't help me at all anyway. I know, due to the fact that my hand will get eventually worse that I may have to have some help ... when you bring aides in here, and I hate it-I absolutely hate it, and [she], I know, doesn't like it. She doesn't want to help, but yet she doesn't want the aide to come in, so where am I at? I mean, something has to change .. J can just feel that she doesn't want to help me but is doing it anyway. For some individuals, work was reported as stressful, but it was a self-imposed stress and therefore perceived as manageable. In summary, most participants reported some stress, but only in the case of the couple having marital troubles was this an overwhelming aspect of their lives. Personal Ideology For some people, difficulties or crises may be managed more effectively if they can turn to some religious or spiritual belief (Tedeschi & Calhoun, 1995b, p. 64). Individuals were asked to indicate whether or not they had a belief in God or some higher power, and how their religious/spiritual beliefs may have changed over time. Couples were examined together in order to identify if their religious or spiritual 129

PAGE 145

ideologies and values were similar. Six of the couples mutually agreed that they had strong religious or spiritual foundations and a firm belief in God or some higher power. One couple agreed that they did not have a belief in any higher power, as it was illogical to both of them. Two couples indicated they had both been raised with strong religious experiences, but had lost their faith as they grew older, although the wives both maintained a firm belief in a higher power. For one couple, the wife felt that her life was in God's hands, although her religious conviction had wavered since her husband's injury: I believe that everything happens because He has a plan for us all God only knows what mine might be, but I believe that I'm doing what I should be doing. It's very strong for me .... It [faith] comes and goes, because I'll get really angry and I'll absolutely try not to think about or even go into a church, and all of a sudden, here I am, you know, it will come back to me. It's ok. Maybe that's why I'm having so much trouble going back to church. It's like my grandma said. "What are you doing? No wonder you're like that. Go to church and everything will be perfect." .. .It can come and go. Right now it's gone. Her husband also had experienced diminished faith in God following his injury: When we were kids ... God was in our life. God was going to look after you, you know, and you would live life doing the right things. Once I got hurt, I was at Craig with a guy and his mom was very, very religious and she was really spiritual. Every time she'd see me she'd say, "God did this to you for a reason." And that's kind of when I lost my faith, I guess. What did I do that was that bad to put me in this chair? I do find it hard to go to church actually like I used to. It's just, I don't know; I have a hard time believing in it now. Another couple believed that religion didn't play an important role in their lives: 130

PAGE 146

I believe that there's something more than us just being out on our own. I have a real problem with some of the organized religions, I'm not sure that they don't do you more harm than they do good. But you have to believe what you believe. I think it's important to try and do unto others, you know. I try not to hurt anyone's feelings or do anything mean and if you can help someone you should try to do that. That's about as far as it goes with the beliefs. (wife) I grew up in a Christian home and went to church and all that kind of good stuff. After I got married I kind of got away from it ... When my mother started getting worse and sick and all that kind of stuff I really turned to it and started praying but nothing seemed to help. I figured I'd give it a try and see what happened for me. My mom was really religious, too, and she got into all this television stuff with all these miracles and putting this stuff on your back. I didn't have the heart to tell her it wasn't working and she was wasting her money She tried it and it didn't work, then for probably the first two or three years I was really praying and hoping and saying "just show me any sign." There was none and I just kind of gave up and it's not for me. (husband) When asked to indicate what was the most important value in human living, the key theme that emerged encompassed sentiments of love, courtesy or respect for others. Most individuals stated that these values of compassion for others were key to living a satisfying life, as described by one wife: Love, respect and honesty-they're all really important and they're all connected. You need one for the other. I think if you have love in your life you also have to have respect for one another and each other's individuality and each other's strengths and weaknesses. Faith is important. There's nothing without honesty, because if everyone was just telling a pack oflies, then that kind of all falls apart, too. They all go together. 131

PAGE 147

Another wife indicated that respect and courtesy for others would make the world a better place: I think we have a tendency to forget that we're speaking to someone else that has feelings and I think 90% of our problems would just stop if we could just be courteous. You don't have to like ... you don't have to agree, but if we treated each other with some compassion it would be a lot different in the world. A husband whose injury required a lot of assistance from his wife and family -tasks that often take a great deal of love and trust in others stated: I think it takes love-and that means for everybody, not just in your family-to help each other and we're all on the Earth together. I think with love, trust and support, anyone can get through anything. Summarizing the personal ideologies and values, it appeared that most couples were generally in harmon y regarding their spiritual beliefs and assertion of human values. For those couples with more disparate viewpoints other stresses in their relationships were apparent and the SCI did not appear to b e the sole factor contributing to stated difficulties. Pre-existing characteristics of the relationship were mentioned as contributing to these stresses, particularl y as reported by the wives. The stories overall created a picture of generally well-adjusted satisfied couples who had established parameters of behaviors. Many of the wives had sought to balance their lives by engaging in outside activities such as exercise classes or work, which 132

PAGE 148

helped modify some of their stresses of caregiving. For those wives whose outlets were more limited, dissatisfaction with the caregiving role were apparent. The husbands, with no exception, expressed admiration and Jove for their wives, even in light of conflicts within the relationship. For some of the men, there was a component of guilt being a burden to their partners, or limiting the recreational and social options in which their partners could engage. These facets of burden or limitation were generally not echoed by the wives. In fact, the wives tended to attribute any caregiving stress more to their own inability to relinquish control over these activities than to the dependency of their husbands. Clear messages from these stories showed the establishment of independence as a turning point, the importance of family, the wealth or lack of social support, and the degree to which the SCI became a part of the context of their lives, but not necessarily the most dominant feature. The emphasis on family may represent a shifting of priorities for these couples who, fo1lowing the injury, chose to focus on different factors in their lives that brought them fulfi11ment. However, there remained a tone of uncertainty for the future in nearly a11 of the stories. Undoubtedly, personality characteristics of the participants came into play that, on a positive note, helped them cope, adapt, and even grow from the experience of SCI. For others, their inherent 133

PAGE 149

characteristics limited their ability to deal with the adversity. The personality type may have been a foundation for the main characters that emerged in these stories. Therefore a final analysis involved assigning one of the McAdams imagoes (agentic and/or communal) described previously in Chapter 2 to each of the individuals. McAdams indicated that imagoes characterize numerous aspects of the individual (McAdams, 1993, pp.127-132): 1. They express our most cherished desires and goals. 2. They personify our traits and recurrent behaviors. 3. The give voice to individual and cultural values. 4. They are often built around significant others. 5. They may signal a fundamental life conflict. Each of the study participants actually had characteristics associated with several imagoes The agentic imagoes appeared in the life stories fairly frequently. For some wives the warrior was evident in their ability to "promote peace and stability through strength" (McAdams, 1993, p. 13 7). The wives often assumed a more dominant role in the relationships having friends as their "allies ," and going to battle with insurance companies, non-supportive family members or the medical community. 134

PAGE 150

The imago of the traveler was not as apparent in the current lives of the participants; however, many had, in their younger days, been far more nomadic. Whether it was the limitation of the injury or generally settling into an established life that limited the exploratory nature of travel remained unclear, but several individuals indicated that travel was a desired but missing aspect in their lives The sage was present in many of the stories, exhibiting the degree to which individuals achieved some degree of power or mastery through knowledge with the men knowing how to manage their own self-care, or the women learning their own coping abilities through adversity and loss. A final agentic imago the maker was also present, although to a lesser degree This imago was apparent in the act of creating something, whether it was through artistic expression, expressing creativity at work or through doing something as simple as domestic chores. Although aspects of the maker were apparent in nearly all the life stories it was a less dominant imago. There was much clearer evidence of communal characters in these life stories. The imago of the lover, for example, was present in many of the couples, with both partners expressing mutual passion for one another, although not necessarily in a sexual context. The shared passion and affection was evident in how close relationships with loved ones, whether spouses or other family members helped shape the individual personalities and behaviors of the study participants 135

PAGE 151

Conversely, a lack of affection or love for close relatives or friends also helped define individuals and their present personalities to a great extent, as exemplified by one woman stating: I grew up hating my father. I had no respect for my father. In my mind he was just the kind of person that I would never marry. I'd never be his friend. When I look at him, I think I'll never be like him and that keeps me straight. Not surprisingly, a significant communal imago in these stories was that of the caregiver. This character was apparent in the individual who not only provided for and gave of himor herself to loved ones, but also one who was willing to sacrifice his or her own needs to those of family and friends. Both husbands and wives expressed characteristics of the caregiver, particularly in regard to caring for children or other family members For one husband, having children was a happy sacrifice: It has an impact on your life as far as what it does to your schedule, where you live for schools, what you do for vacations, what you do when you get home from work. I think they've changed what I like to do. I'm totally fine with my schedule being changed. The things I used to do before I had kids-they're not at all how I spend my free weekends now, and that's fine. Obviously, those acts specific to providing personal assistance would fit with the caregiver imago as well, even in cases where the caregiving was resented. This was apparent in comments by individuals who had been caregivers to younger siblings or to other relatives, as voiced by one of the men: 136

PAGE 152

.. .I kind of became an instant babysitter since I was the oldest. My sister was born when I was almost 13 so then I was the babysitter. I think I was a little bit bitter about it at the time but in those days and age you didn't question those things, so I just went with what I was dealt. Similarly, one of the wives recognized her caregiving had extended to her younger life, and reported: I had a lot of experience in my life as a parental child. I was the oldest, so I tend to dominate or am very opinionated. I'm even the parental child with my mom. My mom definitely needs some help sometimes. I'm the person that if anybody gets in trouble they call me. I bring it on myself, I know I do In short, the caregiving role was not unique to the wives. Several of the men took on characteristics of the caregiver, particularly with regard to relationships with their children. The communal imago of the friend was also apparent in many of the study participants, with reports of close friendships consistently stated in positive, fulfilling terms. In fact for many of the wives, the presence of a good friend someone who was a sounding-board for frustrations and difficulties-was extremely important. The value of friends was also highlighted in the narrative section regarding "significant people," with nearly every participant naming at least one good friend who provided support or encouragement at some point in their lives. 137

PAGE 153

The final communal imago the ritualist, did not appear consistently in the life stories of the study participants. Accounts of preserving traditions that united family and friends were not explicitly stated as being characteristics held by the study participants. However, several of the individuals reported a family life or parental influence that was somehow grounded in traditional domestic or cultural values and activities. These accounts were, for the most part, positive and pleasant memories. Individuals who appeared to be characterized by both agentic and communal imagoes (the healer, teacher, counselor, humanist, arbiter) were not clearly identifiable. It was reasonable to assume, however, that for those who were parents, elements of all these characteristics were present from time to time. A more prominent imago was the person who was both low in agency and low in communion-the escapist or, more particularly, the survivor. In McAdams' categorization, the escapist represents individuals who avoid responsibilities or unpleasantness living solely for diversion and amusement (McAdams 1993 p. 171). While several ofthe stories contained episodes of pleasure-seeking and a desire to relinquish responsibility none ofthe individuals were dominated by this particular imago. The survivor on the other hand was clearly present in the perseverance expressed by individuals in the face of adversity -the ability to keep going regardless of the obstacles. The survivor imago 138

PAGE 154

was often illustrated through the injury itself, or through some kind of redemption after experiencing pain or loss. Several individuals indicated that these losses led them to a better understanding of themselves, of what was truly important, and how they were strengthened by the loss to some degree. Quite possibly,these stories, with the exception of the SCI, would not be much different from those of any other couples today in fact, these appear to be normal stories of normal people who have experienced and, to the best oftheir ability, adjusted to a serious disruption in their lives. They encounter familial issues difficulties and joys of raising a family-and their lives involve work, friends, growth and loss. For the wives, the activities related to caregiving could be stressful at times; however, it appeared that many of the stresses were self-imposed as these women felt unwilling to or limited in their ability to relinquish responsibility for at least some of the caregiving tasks. The ability of these women to engage in activities independent of their husbands seemed to help moderate the perceived stresses or burdens related to careg1vmg. The Life Story interview process, in addition to providing valuable information to help the investigator explore the issues potentially related to caregiver stress, depression and burden, also enabled the participants to reflect back on their lives-139

PAGE 155

both before and after the SCI. The stories provided information in the form of numerous "snapshots" of key events and people that have helped shape them as the people they are today. The telling of these stories was undoubtedly influenced by the participants' current thoughts and events. For those with current health problems, there was a focus on health issues. For others, family concerns were at the forefront of their Stories. Nonetheless, when viewing these stories as a whole, a clear picture emerges that includes pride in independence and a need to be in control of one's self and surroundings, the impm1ance of family and friends, and, in most cases, mutual love and respect for one another. In the occasional absence of such love and respect, there still was a clear indication of concern and compassion for the spouse's well being. 140

PAGE 156

CHAPTER SIX DISCUSSION The findings from this study reinforce the concept that a spinal cord injury impacts not only the individual incurring the neurologic trauma, but also the family as a whole. This, alone, is not a new finding, as the broad impact of SCI on spouses has been described previously in detail (Weitzenkamp et al., 1997; Kreuter et al., 1998; Kreuter, 2000; Chan, Lee & Lieh-Mak, 2000). This study is unique however, in demonstrating that the adjustments to SCI continue over the span of the relationship, and even for those couples married for many years post-injury, there are stresses and concerns on the part of both members of the couple. When comparing the couples on measures of stress and depression, the men with SCI were significantly more depressed than their spouses, suggesting that the adaptation to SCI takes many years and depressive symptomatology may not resolve quickly. Of note, data from earlier studies of marriage and SCI indicate that couples married after the injury reported greater life satisfaction and better adjustment (Crewe, Athelstan, & Krumberger, 1979; Crewe et al., 1979; Crewe & Krause, 1992). In these studies, no ratings of the spouse's life satisfaction or adjustment were examined, however. The only comparable study was conducted in Great Britain with an older sample that 141

PAGE 157

included individuals married both before and after the injury, and who had lived with SCI for a considerably longer time (Weitzenkamp et al., 1997). In that study, the spouses reported greater depressive symptomatology than their injured partners. This may suggest that with the passage of time, individuals with SCI adapt or adjust better to their injuries, while their spouses experience increasing depression, possibly related to declines in their own health and abilities as they, themselves, grow older. Findings from the qualitative interviews in the present study suggest that the husbands with SCI often felt guilty for being a burden to their wives and children, even if this sentiment was not shared by their spouses. The qualitative information also indicates that wives are less likely to complain to others, or ask others for assistance, and this may also contribute to increasing difficulties as time passes. The burden of care in this study was defined as the number of hours spent per week providing assistance to the spouse with SCI. There was a clear, yet non-significant pattern of greater stress and depressive symptomatology in the wives who spent more than ten hours a week in caregiving activities. They also reported significantly higher burden scores. This finding was not fully consistent with previous studies of caregiver burden, which found no direct connection between perceived burden and hours of care provided (Arai, Masui, Sugiura, & Washio, 2002; Arai & Ueda, 2003). Notably, these studies examined burden in a very small sample offamily caregivers of 142

PAGE 158

elderly individuals, exploring longitudinal changes in perceived burden (using a different burden inventory) and hours of care. The authors suggested that the absence of a relationship between care giving and perceived burden might relate to the definition of burden. In that study, the amount of paperwork needed to coordinate outside services and the related costs of these services contributed to a sense of burden on a different level. In the present study, it is possible that burden scores were higher for the wives who provided more assistance because the type ofhelp needed for people with SCI can be very time consuming and physically intensive. For example, performing a bowel program for a fully dependent individual may occur three to four times a week and take several hours each time, including insertion of suppositories and/or performing digital stimulation to increase peristalsis and relax the external anal sphincter, manual evacuation of stool from the rectum, clean up, dressing, and transferring into a wheelchair. From the qualitative interviews, both husbands and wives voiced concerns that they did not want outside caregivers to perform intimate tasks such as bowel care or changing catheters. In addition, there were few instances ofavailable outside help, leaving the wives with limited resources to share the load of caring for their husbands. The finding that only the Emotional Burden subscale score of the CBI was not significantly higher for wives providing more hours of care suggests that the perceived burden is a function of time and effort rather than psychological distress. 143

PAGE 159

The findings from this study also indicate that older caregiving wives experienced greater perceived stress and social burden than younger wives. This finding is not consistent with the Alzheimer's literature (Fitting et al., 1986; Zaritet al., 1980), which suggested that younger caregivers experience worse psychosocial outcomes than older caregivers. The only study that examined long-term caregiving spouses and SCI (Weitzenkamp et al. 1997) did not explore differences in ages of caregivers, so no directly comparative data are available. It is likely that in the case of Alzheimer's disease, younger caregivers experience greater psychosocial distress because they are facing the continual cognitive decline and ultimate death of their loved ones and they may not be emotionally prepared to anticipate a life without that partner; whereas the older spouse may be more accepting of this demise and their own decline and death over a shorter time span. The present study found the stresses of caregiving more likely to take a toll on older wives, due to the physical demands and the social limitations imposed by the SCI. The older wives may find it more physically difficult to continue providing assistance with transfers, dressing, bathing and other strenuous activities, or, doing these activities safely may take more time. Their ability or willingness to engage in outside activities may thus be limited by their own fatigue. 144

PAGE 160

Even though many of the differences between younger and older wives of men with SCI were not statistically significant, there was a consistent trend of the older wives having worse scores on all measures of stress, depression and burden. These data, although cross-sectional, suggest that as their partners age and require more assistance, the likelihood increases that the wives will have worsening psychosocial outcomes related to caregiving. Supporting this is the concern voiced in the qualitative interviews about the partner's declining health as he grows older; in fact, both individuals in the oldest couple involved in the Life Story interviews were already experiencing health problems related to aging. Nonetheless, an important mediating factor that helped diminish stress and depressive symptomatology was the availability of social support and a sense of control, similar to findings from research by Decker and colleagues (1990). Attachment theory was used to explore differences in stress, depression and burden scores among the wives of men with SCI. Although we hypothesized that wives in those couples with similar attachment styles would have less stress, depression and burden, the study found no significant differences except for higher scores on the Emotional Burden subscale of the CBI. This finding, which contradicts the hypothesis, in fact makes sense. In this sample, most ofthe couples had a "close," "dependent," or "close dependent" attachment style. It is quite likely that couples 145

PAGE 161

who share more intimacy and interdependence would have greater concern for one another, and greater sensitivity to the needs of each other. As a result, the perception of emotional burden could be higher. For example, a wife may indicate she was embarrassed by or resentful of her spouse because she perceived that any actions or difficulties on his part were a negative reflection on her supportive, caregiving abilities. There was also a trend toward greater stress and depressive symptomatology in wives of more closely attached couples. There is evidence in the literature that depressive symptoms in one spouse influence those of the partner, pat1icularly in close marriages where distress in one spouse is distressing to the partner (Tower & Kasl, 1995). Conceivably, individuals in more avoidant relationships do not have the emotional investment in their partners, and are less likely to experience distress due to difficulties their partner is facing. Because the proposed hypotheses were not uniformly supported by the study results regression analyses helped clarify some of the findings, providing other explanatory variables for the noted outcomes in the female partners. Stress and depression in the wives were closely interrelated, with higher scores on one measure associated with higher scores on the other. In addition poorer self-perceptions of health were related to greater depression and physical burden, while perceptions of the spouse's health were unrelated to any ofthe selected outcomes. Others have described the negative 146

PAGE 162

health effects of caregiving (Kiecolt-Glaser et al., 1991; Kiecolt-Glaser et al., 1995; Kiecolt-Glaser et al., 1996; Schulz & Beach, 1999; Vedhara et al., 1999; Beach, Schulz, Y ee, & Jackson, 2000; Glaser et al., 2000; Vitaliano et al., 2002; Lee, Colditz, Berkman, & Kawachi, 2003); however, the present study also highlights the linkage between perceived health and depressive symptomatology and perceived physical burden. These data underscore the importance of initiating and continuing the process of educating caregivers to care for themselves with as much intensity as they care for their partners with SCI. While the quantitative findings identified trends in the selected outcomes or significant associations between the outcomes and their related factors, it was only through the qualitative interviews that new, useful information was obtained. The narrative stories focused on issues not captured by the quantitative instruments-such as family and past history, significant people and personal ideology-adding a valuable layer in our understanding of caregiver outcomes. Conclusions from this study must be drawn with caution. The study sample was relatively small, and some of the quantitative analyses were not adequately powered due to missing data that, at times, reduced the valid sample size. The absence of consistent significant findings to support the study hypotheses may be due to the selected instruments used to assess depression, stress and burden. The CES-D and PSS may not be adequate to capture the nuances of caregiver distress in this unique 147

PAGE 163

population. The study also is somewhat limited by the sample selection. The patticipants could not, by definition, reflect the viewpoints of couples who have lived with SCI for a shorter period of time, nor couples married after the SCI, nor those who divorced after the SCI. In addition the qualitative data reflect only those sentiments of the selected participants and may not be representative of couples with SCI in a broader context. Nonetheless, the findings from this study provide some insight to the stress process model described earlier in this thesis, and enable us to more specifically identify the actual variables that serve as mediators in this model. The model, which was originally based on Alzheimer's caregiving, may now be revised to illustrate factors related specifically to SCI caregiving, as shown in Figure 4. This new model like the original, is comprised of four domains. The first the backgi ound and context of distress, includes not only the socioeconomic characteristics, caregiving history, elements of childhood experiences, and composition of the family and network but more specifically focuses on the existing stability of the relationship. This is evident in the life stories in this study that describe the pre-injury characteristics of both the husbands and wives and the strengths and weaknesses of the marital relationship. The second domain includes 148

PAGE 164

both primary and secondary stressors as well as Intrapsychic stressors and describes the stressors encompassed by the demands of caregiving. These are defined from the present study as the "load" of caregiving (number of hours providing assistance, dependency of the injured partner, and the caregiver's self-perceived health). Also included are the secondary stressors of role strains (work/caregiving conflicts, financial concerns and social life limitations), and intrapsychic strains (perceived burden, and loss of control over one'stime, self and needs). The third domain mediators of stress, includes social support and positive self-esteem or sense of control Fim'tlly, the fourth domain represents the outcomes of stress, which include stress and depression, as well as concerns with physical health. Figure 4 Proposed SCI Caregiver Stress Model ......................................................................................... : ........................................... . . : 11 ir--, .. T 1 re/atio;ship statit $ 1 -j ,,,.,,!,, = I Family -histofv ll [:_ Number of ca regiVIng 1 Work confli_cts i Perceived l S : hours Financiiil b d l tress i 1 ur en l Depression I ; j : ,< Lc ____ --I 1 -.> ._, I i j _-I --. health : .,, --:I d e pendenCJ' I J l I I L. ____ i _____ i _____ j _____ i _____ i ____ ,j + + Perceived social ._ t support :Positive self. -::,-evaluation: '-;, 149

PAGE 165

In this revised model, the impact of the mediators (perceived social support, perceived self-evaluation) on the secondary strains and outcome is no longer unidirectional. The proposed model is more fluid, suggesting that there is a continuous flow in which conflict, strain, burden, and loss of control mediate and are mediated by the positive aspects of social support and self-evaluation. For example, a spouse who is experiencing a lack of support from friends and family may encounter greater stress in the caregiving role and may become more isolated in that role. In a somewhat circular manner, her increased, often self-imposed isolation, could limit opportunities to strengthen relationships with friends and family that would help build a foundation of support structure. On the other hand, a stronger social support structure could more likely provide a positive beneficial impact for the caregiving spouse, thus alleviating some of the stress and sense of burden. This was evident in the stories of women who allowed time for themselves to engage in social activities with their friends or through their jobs. Another important aspect of this model is the perception of one's own self and strengths. The qualitative interviews from this study suggested that those women who described self-reliance or an achievement of independence in some aspect of their lives were better able to cope with the demands of caregiving, while still maintaining time for their own needs. Conversely, those who did not voice particular events or 150

PAGE 166

activities that demonstrated their own independence in the past were more likely to report an inability to put their own needs first while still maintaining a positive caregiving role. While this model is perhaps more fragile due to the bidirectional impact of stressors and mediators, it appears to better represent the stress processes encountered by wives of men with SCI, incorporating aspects of their lives both before and after the injury. It is, in fact, a more flexible model, encompassing a wider range of elements and providing more explanatory strength. In addition, this model illustrates some of the specific issues highlighted by the study participants themselves, and provides a basis from which to develop further studies of caregiver outcomes, the possible content of which is discussed in the following chapter. 151

PAGE 167

CHAPTER SEVEN CONCLUSIONS The purpose of this study was to explore the impact of SCI on female spouses whether or not they provided personal assistance services to their injured partners. Four research hypotheses were tested. The first hypothesis was that partners of men with SCI would report greater stress and depression than their injured partners. This hypothesis was not supported. The husbands with SCI reported significantly higher scores on the CES-D than did their wives although stress scores were not different. The second hypothesis was that partners of men with SCI who provided more than I 0 hours per week of assistance would report greater stress, depression and burden than those providing 10 or fewer hours of assistance per week This hypothesis was partially supported. Those wives providing more than I 0 hours of assistance did report significantly greater burden in all areas except for the Emotional Burden subscale There were non-significant trends toward greater depression stress and emotional burden as well. The third hypothesis was that older partners of men with SCI would report less stress depression and burden than younger partners The data did not support this I 52

PAGE 168

hypothesis, but showed that older partners had significantly higher scores on measures of stress and social burden than younger partners. The final hypothesis was that in couples with greater similarity of attaclunent style, the female partners would report less stress, depression and burden than in those couples with less similar attachment styles. The data did not support this hypothesis and the CBI Emotional Burden subscale score was significantly higher for the wives in couples having more similar attaclunent styles. There were non-significant trends toward greater depressive symptomatology, greater stress, more physical burden and more social burden in these women, as well. The various regression models provided additional insight into factors associated with stress, depression and burden, suggesting that self-perceived health, duration of injury, and the perception of difficulties in other areas (such as bumout, financial concems, trouble sleeping, etc.) contributed to worse outcomes. The R2 values for the models, however, suggest that additional, unexplored factors are likely contributing to scores on the various outcome measures. A strength of the present study was the addition of a qualitative exploration into the lives of couples living through the experience of SCI after mmTiage. Their 153

PAGE 169

compelling accounts of high, low and turning points in particular, highlighted the importance of family, support from friends, and a pride in achieving independence in various arenas. Based on these findings, there appears to be compelling evidence that spouses not only play an integral role in the well-being of their injured partners, but that they, too, have issues and concerns that need to be addressed. While this type of family involvement is a feature of most SCI rehabilitation programs, there is an on going, long-term justification for intervention. The value of being able to identify assistive services, respite options or simply a recognition that the spouse also has needs, may help allay some of the fears and frustrations of caregiving. However, it is not caregiving, alone, that is the nidus for these frustrations and stresses. In fact, it has been suggested that living with a person who has a disability is, in itself, stressful (Weitzenkamp et al., 1997). These stressors may be due to fears for the long-term health and well-being of one's partner with a disability, or due to social limitations imposed by the disability. The wealth of infmmation provided through the qualitative interviews helped describe and define some of the stressors more clearly than a quantitative measure could. The stories told by these individuals, and the information learned in this investigation, suggest there is much yet to be learned about the nature of caregiving in SCI and the impact of such an injury on the family. A more detailed understanding of the 154

PAGE 170

interplay between health factors and psychosocial outcomes in long-term partners living with SCI is warranted. The hypotheses examined in the present study should be explored with other samples of couples with SCI including those who married after the injury, and couples in which the female partner has the SCI. Another avenue of investigation would focus on the persons with SCI, and the degree to which they perceive themselves as being burdens to their loved ones. In addition a prospective, longitudinal study would help explore the factors that keep marriages together-this study could not explore the factors that lead to divorce following SCI, which is a significant problem. Furt}?.er studies of adaptation to physical trauma, for both the individual personally experiencing the trauma and for his or her loved ones, would help highlight characteristics that individuals use in coping, reprioritizing and ultimately growing from the adversity. With regard to attachment theory, additional studies of adult attachment and caregiving, with a more detailed focus on differences between attachment styles and the factors contributing to those styles may provide additional insight into the relationships of spousal caregivers and their partners with disabilities. These suggested avenues of inquiry may help clinicians working with couples facing a new SCI better preparing them for the difficulties they may experience after returning home to an altered life with disability. Couples having lived many years with SCI may also benefit by the experiences of others who have shared their stories, pointing out successes and difficulties of life with SCI. 155

PAGE 171

Finally, this investigation has provided new and unique insights into the complex nature of relationships following SCI. Health and service providers can use this information to help design and develop intervention programs to help families cope more effectively with the consequences of SCI in order to help couples maintain strong, healthy and happy relationships. 156

PAGE 172

APPENDIX A Data Collection Instruments 157

PAGE 173

CENTER FOR EPIDEMIOLOGIC STUDIES DEPRESSION SCALE (CES-D) Please think back over the past week, and indicate how many times (stated as days in the week) you have experienced the particular symptoms listed below. 0 1 2 3 4 5 6 7 1. I was bothered by things that usually don t bother me. 2. I did not feel like eating; my apl'_etite was poor. .., I felt that I could not shake off the blues even with .), help from my family or friends 4. I felt that I was just as good as other people 5. I had trouble keeping my mind on what I was doing 6. I felt depressed. 7. I felt that everything I did was an effort. 8. I felt hopeful about the future. 9. I thou_ght my life had been a failure. 10. I felt fearful. 11. My sleep was restless. 12. I was happy. 13. I talked Jess than usual. 14. I felt lonely. 15. People were unfriendly. 16. I enjoyed life. 17. I had crying spells. 18. I felt sad. 19. I felt that people dislike me. 20. I could not get "going". 158

PAGE 174

PERCEIVED STRESS SCALE The questions in this scale ask you about your feelings and thoughts during the last month In each case you will be asked to indicate. how often you felt or thought a certain way. Although some of the questions are similar, there are differences between them and you should treat each one as a separate question. The best approach is to answer each question fairly quickly. That is, don't try to count up the number of times you felt a particular way, but rather indicate the alternative that seems like a reasonable estimate. Never Almost SomeFairly Very Never times Often Often I. In the last month, how often have you been upset because of something that happened unexpectedly? 2. In the last month, how often have you felt that you were unable to control the important things in your life? 3. In the last month, how often have you felt nervous or "stressed"? 4 In the last month, how often have you dealt successfully with irritating life hassles? 5 In the last month how often have you felt that you were effectively coping with important changes that were occurrino in your life ? 6. In the last month, how often have you felt confident about your ability to handle your personal problems? 7 In the last month, how often have you felt that things were going your way ? 8. In the last month, how often have you found that you could not cope with all the things that you had to do? 9. In the last month, how often have you been able to control irritations in your life? 10. In the last month how often have you felt that you were on tOll_ of things? 11. In the last month, how often have you been angered because of things that happened that were outside your control? 12. In the last month how often have you found yourself thinking about things that you have to 13. In the last month, how often have you been able to control the wav you spend your time? 14. In the last month, how often have you felt difficulties were piling up so high that you could not overcome 159

PAGE 175

The Revised Adult Attachment Scale (Collins & Read) Please rate the e xtent to which each of the following statements des cribe s your feelings on a scale from 1 (not at all characteristic) (very characteristic) Not at all Very Characteristic Characteristic I. 1 find it relatively easy to get close to others 2 4 5 .) 2 1 am comfortable depending on others 2 3 4 5 3 1 am comfortable having others depend on me 2 3 4 5 4. I know that others will be there when 1 need them 2 3 4 5 5. 1 do not often worry about being abandoned 2 3 4 5 6. 1 do not often worry about someone getting too close to me 2 3 4 5 7. 1 am somewhat uncomfortable being close to others 2 3 4 5 8. 1 find it difficult to trust others completely 2 3 4 5 9. 1 find it difficult to allow myself to depend on others 2 3 4 5 10. 1 am nervous when anyone gets too close 2 3 4 5 11. Often, my love partner wants me to be more intimate than I feel comfortable being 2 4 5 .) 12. 1 find that others are reluctant to get as close as I would like 2 3 4 5 13. I am not sure that 1 can always depend on others to be there when 1 need them 2 3 4 5 14. 1 often worry that my partner doesn't really love me 2 3 4 5 15. 1 often worry that my partner won't want to stay with me 2 3 4 5 16. 1 want to merge completely with another person 2 4 5 .) 17. My desire to merge sometimes scares people away 2 3 4 5 18. People are never there when you need them 2 3 4 5 160

PAGE 176

The Caregiver Burden Inventory receiver I expected that things would be different at this point in life efforts aren't appreciated by others in my 161

PAGE 177

PERSONAL SUPPORT/ASSISTANCE QUESTIONNAIRE 1 How old are you? 3. What is your highest level of education: __ elementary/middle school __ high school or GED __ some college __ university (Bachelors degree) __ education beyond university level 3 What is your relationship to your partner: __ spouse __ lover/significant other 4. Do you live in the same house as your partner? __yes no 4 How many other people live with you in your house? (fill in the number in each of the categories below as well as their ages) parents Age(s) : __ children Age(s): __ others: Age(s): __ 6. How long have you known your partner? __ years 7. During a typical day, how many hours do you spend with your partner? __ hours/weekday __ hours per weekend day 8. How would you rate your partner's health now? __ excellent __ good fair __ poor 9. How would you rate your own health now? __ excellent __ good fair __ poor 10. Sometimes assisting your partner involves the activities you would be doing anyway, like helping with laundry or cooking. It may also involve additional activities because of the injury. How many hours per day (on a typical day) do you provide additional assistance to him because of the spinal cord injury? __ hours per weekday __ hours per weekend day 162

PAGE 178

PERSONAL SUPPORT/ASSISTANCE QUESTIONNAIRE Page 2 1 1 How long have you been providing this additional type of assistance to your partner? ________ years 12. Are you paid or otherwise compensated (non-monetarily) for providing assistance? ____yes no 13. Are you the only person who usually provides assistance to your partner? ____yes no If "no," how many hours of assistance are provided by someone else? __ hours/weekday __ hours per weekend day Who is/are the other person(s)? ------------His/her/their relationship to your partner : __________ 163

PAGE 179

PERSONAL SUPPORT/ASSISTANCE QUESTIONNAIRE Page 3 14. What specific activities do you do to help your partner because of his spinal cord injury. Please indicate how much time is spent in each activity using the following scale: 1 = < 1 hour (0-59 minutes) 2 = 1 to 2 hours (60-120 minutes) 3 = 2 to 5 hours 4 = 5 to 10 hours 5 => 10 hours (601+ minutes) (121-300 minutes) (301-600 minutes) Be sure to include personal assistance, household management or other chores that you do because your partner needs assistance or is no longer able to do these activities. Activity EXAMPLE: bowel program EXAMPLE: getting dressed in morning Amount of time 2 #times per week 3 7 15. Has the amount of assistance you provide to your partner changed over the past three years? no ___yes lf"yes," how has it changed: __ more time needed to provide help __ more activities need help __ more difficult to provide help __ partner wants more help __ other: Specify 164 less time needed __ fewer activities need help less difficult __ partner wants less help

PAGE 180

PERSONAL SUPPORT/ASSISTANCE QUESTIONNAIRE Page4 1 6. Many people experience a variety of physical and emotional consequences related to having a partner with a disability, particularly when that partner needs extra help. Please check which of the following you have experienced as a result of your partner's disability. __ physical stress or strain emotional stress or strain __ feeling that helping or caring for your partner is inconvenient burnout financial concerns __ fatigue __ anger/resentment __ dissatisfaction with your life __ trouble sleeping, or changes in your sleep patterns __ changes in your own personal plans and goals __ isolation, loneliness fear for the future __ concerns about upsetting behavior (e.g., because of incontinence, memory, etc.) __ interference with other demands on your time __ changes in your outside job, or ability to do that job __ changes in your social life __ having less free time or time to yourself __ feeling overwhelmed 17. Life experiences may cause us to think about things in different ways, or teach us new lessons. Looking back on your life, and knowing what you do now, is there anything in your life since your partner's injury that you would change? __yes no If yes, please explain:-------------------165

PAGE 181

The Life Story Interview (note, the entire interview is shown here. Only selected elements, which are highlighted, were used in the present study) Dan P. McAdams, Northwestern University Revised 1990 Introductory Comments This is an interview about the story of your life. We are asking you to play the role of biographer of self --to construct for us the story of your own past, present, and what you see as your own future as if you were putting together your own biography. People's lives vary tremendously, and people make sense oftheir own lives in a tremendous variety of ways As social scientists, our goal is to collect as many different life stories as we can in order to begin the process of making sense of how people make sense of their own lives. Therefore, we are collecting and analyzing biographies of "normal" adults from all walks of life, and we are looking for significant commonalities and significant differences in those life stories that people tell us. There are a few things about biographies that you should keep in mind during this interview. First and foremost a biography is selective. A good biography tells a convincing story about the individual without burdening the listener with thousands of details. Therefore, it is not necessary for you to tell us everything that has ever happened to you. We are interested in a few key events a few key relationships, a few key themes which recur in your life story. You should concentrate on material in your own life that you believe to be important in some fundamental way --information about yourself and your life which says something significant about you and how you have come to be who you are. Second, a good biography is somewhat objective and analytical. This means that your life story should be conveyed to us in as unbiased way as possible with a critical eye towards trying to make sense of your own life rather than glorifying it. Good biographies are not like eulogies or songs of praise. Rather than extolling all of your virtues we would rather that you tried to give us a sense of what makes you unique as a human being, what aspects of your own experience render your life different from others. This means that you may have to analyze yourself a little bit as you describe your story Thus, we would rather you not tell us about how great (or how bad) your life is. Instead, try to tell us about your unique experience as a human being living at this time and in this place. Finally, it is only fair that we make clear to you our purposes in asking you to construct your biography. Essentially, we wish to catalogue people's life stories so that we may eventually arrive at some fundamental principles of biography as well as ways of categorizing and making sense of life stories constructed by healthy adults living at this time in history and in this place. We are not interested therefore, in pathology, abnormal psychology, neurosis and psychosis. We are not trying to figure out what is wrong with you. Nor are we trying to help you figure out what is wrong with you. The interview 166

PAGE 182

should not be seen as a "therapy session." This interview is for research purposes only, and its sole purpose is the collection of data concerning people's life stories. The interview is divided into a number of sections. In order to complete the interview within, say, an hour and a half or so, it is important that we not get bogged down in the early sections, especially the first one in which I will ask you to provide an overall outline of your story .The interview starts with general things and moves to the particular. Therefore, do not feel compelled to provide a lot of detail in the first section in which I ask for this outline. The detail will come later. I will guide you through the interview so that we can finish it in good time. I think that you will enjoy the interview. Most people do. Questions? [The interviewer may wish to ask for clarifications and elaborations at any point in this section, though there is a significant danger of interrupting too much. If the subject finishes in under 15 minutes, then he/she has not said enough, and the interviewer should probe for more detail. If the subject looks as if he/she is going to continue beyond half an hour. then the ifderviewer should try (gently} to speed things along somewhat. Yet, you don't want the subject to feel. 'rushed" (It is inevitable, therefore, that some subjects will run on too long.} This is the most open-ended part of the interview. It has the most projective potential. Thus, we are quite interested in how the subject organizes the response on his or her own. Be careful not to organize it for the subject.} II. Critical Events Now that you have given us an outline of the chapters in your book --a Table of Contents, as it were --we would like you to concentrate on a few key events that may stand out in bold print in the story. A key event should be a specific happening, a critical incident, a significant episode in your past set in a particular time and place. It is helpful to think of such an event as constituting a specific moment in your life which stands out for some reason. Thus. a particular conversation you may have had with your mother when you were 12-years-old or a pmticular decision you made one afternoon last summer might qualify as a key event in your life story .These are particular moments set in a particular time and place, complete with particular characters, actions, thoughts. and feelings. An 167

PAGE 183

entire summer vacation it very happy or sad or very important in some way --or a very difficult year in high school, on the other hand, would not qualii)t as key events because these take place over an extended period of time (they are more like life chapters.} Jam going to ask you about 8 specific life events. For each event, describe in detail what happened, where you were, who was involved, what you did, and what you were thinking and feeling in the event. Also, try to convey what impact this key event has had in your life story and what this event says about who you are or were as a person. Did this event change you in any way? Jf yes, in what way? Please be very specific here. Questions? [interviewer should make sure that the subject addresses all of these questions especially ones about impact and what the experience says about the person. Do not interrupt the description of the event. Rather askfor extra detail, if necessary, after the subject has finished initial description of the event.} >Ath.onTWho llat lmtna.ct hastlle ha&
PAGE 184

[This kind of redundancy in inevitable.} Event #4: EARLIEST MEMORY Think back now to your childhood, as far back as you can go. Please choose a relatively clear memory from your earliest years and describe it in some detail. The memory need not seem especially significant in your life today. Rather what makes it significant is that it is the first or one of the first memories you have. The memory should be detailed enough to qualify as an "event." This is to say that you should choose the earliest (childhood} memory for which you are able to identify what happened, who was involved, and what you were thinking and feeling. Give us the best guess of your age at the time of the event. Event #5: IMPORTANT CHILDHOOD MEMORY Now describe another memory from childhood, from later childhood, that stands out in your mind as especially important or significant. It may be a positive or negative memory. What happened? Who was involved? What did you do? What were you thinking and feeling? What impact has the event had on you? What does it .say about who you are or who you were? Why is it important? Event #6: IMPORTANT ADOLESCENT MEMORY Describe a specific event from your teenaged years that stands out as being especially important or significant. Event #7: IMPORTANTADULT MEMORY Describe a specific event from your adult years (age 21 and beyond} that stands out as being especially important or significant. 169

PAGE 185

Event #8: OTHER IMPORTANT MEMORY Describe one more event, from any point in your life, that stands out in your memory as being especially important or significant. Choose what you now consider to be the most significant or memorable moment in your life --but one that you have not yet discussed in this section. {After descriptions of the four] Do you have any heroes or role models in your life? Explain. IV. Future Script Now that you have told me a little bit about your past and present, I would like you to consider the future. As your life story extends into the future, what might be the script or plan for what is to happen next in your life? I would like you to describe your overall plan, outline, or dream for your own future. Most of us have plans or dreams which con'cern what we would like to get out of life and what we would like to put into it in the future. These dreams or plans provide our lives with goals, interests, hopes, aspirations, and wishes. Furthermore, our dreams or plans may change over time, reflecting growth and changing experiences. [interviewer will probably have to probe for details here. Get as much specific information as possible.] 2. There are many different ways in which we can be creative in our lives. By creativity here, we are referring not only to artistic expression but also to actions of any sort in which we "give birth" to something, in which we "make something" or "produce something" which exists as our creation, no matter how humble. In what ways, if any, does your plan for the future enable you to be creative? Explain. 3. In what ways, if any, does your plan for the future enable you to make some kind of contribution to others? V. Stresses and Problems i ssu es; problems i6 he-solved,. [ periol:fi : 9fgn;:at stress. Vje you to 9ft1)eselio\V: 170

PAGE 186

yotlr are e xpienenc diffi u.-. nr ....... For e : he t1Te2 st ouih ........ e ........ _"'"'''" ... VI. Persona/Ideology Now 1 would like to ask a few questions about your fundamental beliefs and values Please give some thought to each of these questions .. -In what ways, if any, are your beliefs different from those held by most other people you know? r ... .. [Note: Having no religious beliefs constitutes a religious belief system in itself The interviewer should probe here to understand the nature of this "unbelief' and what, if anything, the subject substitutes for conventional religious beliefs The focus in this section on "Persona/Ideology" should be on those beliefs that differentiate the subject jiom others. What makes him or her unique?} 5. Do you have a particular political orientation? Explain. VII. Life Theme ->.: helpcJ11e understan'_ our __ __ Looking back over your entire life story as a book with chapters, episodes, and characters, can you discern a central theme, message or idea that runs throughout the text? What is the major theme of your life? Explain. 171

PAGE 187

VIII. Final Questions (re: Generativity) I In what ways do you see yourself contributing to the well-being of other people especially younger people of "the next generation"? [If subject is zero in here on teaching as one ofthese.lfsubject is community volunteer, zero in on this If subject is neither of these then leave it completely up to the subject to answer questions in ways that he or she deems fit. If subject sees absolutely no way in which he or she makes a contribution (even after interviewer probes further), then the subject cannot answer following questions, and the interview comes to an end.} 2. Describe this contribution in detail. What do you do? Give specific examples. 3. How did you develop this interest or activity? 4. Why do you think you do this? 5 Who has influenced you most in this regard? 6. Is there anything from your childhood or your past that helps us understand why you make this kind of contribution to others? 7. How does this contributing activity fit into the life story you have already described for me? [Probes needed, most likely.] 8. What kinds of rewards do you receive for your contributions? 9 What kinds of sacrifices do you have to make? What are the negatives? What problems do you experience in this? I 0. How do you think that society looks upon you in this regard? How do your friends and acquaintances look upon you? Your family? II. What might change in your life to help you to do this work better? What might change in your life that would keep you from doing this work any more? 12. What else do you think I should know about this contribution you make? IX. Other 172

PAGE 188

APPENDIXB IRB Approval Letters 173

PAGE 189

University of Colorado at HUMAN SUBJECT RESEARCH COMMITTEE University of Colorado at Denver Campus Box 129, P.O. Box 173364 .Denver, CO 80217-3364 DATE: TO: FROM: SUBJECT: MEMORANDUM April20, 2001 Susan Cablifue, M.A. Deborah Kellogg, HSRC Chair Human Subjects Research Protocol #702 Panners of Men with Spinal Cord Injury: Issues Related to Providing Assistance" The University of Colorado at Denver's JRB, the Human Subjects Research Committee, will accept the protocol and approval granted by HealthONE AJiiance Institutional Review Board. Their deadline date and a]) of their procedures for maintaining data and signed consents will apply. Should the research extend beyond their review deadline, please provide us with the re-approval documents. Keep us informed regarding any changes in procedures or research design as well as any human subjects problems that might arise (303.556.4060). Good Luck with your research. 174

PAGE 190

GD HEALTHONEALLIANCE August 20, 2003 Susan Charlifue, MA 342S S Clarkson Street Englewood, Colorado 80110 Dear Ms. Charlifue; !Nsrm.IT!ONAL REVIEW BOARD As Chainnan of the Health ONE AUiance Institutional Review Board I am pleased 1o inform you that. research protocol: "Partners of Men wJtb Spinal Cord Injury: Issues Related to Proving Assistance" was reapproved for one year on August 20, 2003 by the HealthONE Alliance IRB. No changes are required. This study is closed 10 enrollment. The HealthONE Alliance IRB requires a one year report on all approved projects, or if completed in less than one year, a final report atthe termination of the project. It is your responsibility after cine Year to either request reapproval or to advise the Board of study termination. In addition, it is essential. that all changes in protocol, adverse events, complications or unanticipated outcomes be reported to this Board promptly. Very truly yours; C1 ..... ... =- J t ,.. ..... -.. Sami Diab, MD Chairman HealthONE Alliance Institutional Review Board 899 Logan Streec I Suite ::103/ Denver, Colorado 80203 3 154 I Ph. 303.996.0863 I Fax 303.996.0870 175

PAGE 191

HEALTH ONE ALLIANCE August 21,2002 Susan Cbarlifue, MA 3425 S. CJarlcson Street Englewood, Colorado 80110 Dear Ms. Charlifue; INS11TUT10NAL REVIEW BOARD As Chaimum of the HealthONE Alliance Institutional Review Board I am pleased to infonn you that researchprotocol and infonned consent (dated 3127/02): ''Partners or Men with Spinal Cord Injury: Issues Related to Proving Assistance" were reapproved for one year on August 21, 2002 by the HealthONE Alliance JRB. No changes are required. The HealthONE Alliance IRB requires a one year report on all approved projects, or if completed in less than one year, a final repon at the tennination of the project. It is your. responsibility lifter one .year to either request reapproval or to advise the Board of study termination. In addition, it is essential that all changes in protocol, adver.;e events, complications or unanticipated outcomes be reponed to this Board promptly. Very truly yours; Jeffrey V. Matous, MD Chairman HealthONE AJJiance Institutional Review Board I 850 Hiah Strr-ct/ Otnver, Colorado 80J Ill/ Ph. 303.839.68ll I Fax 303.839.6836 176

PAGE 192

@) HEALTHONE ALLIANCE August 15, 2001 Susan Charlifue, MA Craig Hospital 3425 S. Clarkson Street Englewood, Colorado 80110 Dear Ms. CharUfue: INSTITUTIONAl IUVJEW BOARD As Chairman of the HealthONE Alliance Institutional Review Board J am pleased to inform you that your research protocol and informed consent for: "Partners of Men with Spinal Cord JJUury: )&sues Related to Providing Assistance" were reapproved for one year on August 15, 2001 by the HealthONE Alliance IRB. The HealthONE Alliance IRB requires an annual report on all approved projects. It is your responsibility after one year to either request reapprovaJ or to advise the Board of study termination. In addition, it is essential that all changes in protocol, adverse events, C()JJlpJications or unanticipated outcomes be reported to this Board promptly. Very truly yours; Jeffrey V. Matous, MD Chairman HealthONE AJiiance Institutional Review Board 1850 Hl11h S.,....,1/ Dr"""' Colorado 80218/ Ph. 303.839.6811/ Fox 303.839.6836/www.hehhi.OT@ 177

PAGE 193

August 9, 2001 Susan Charlifue, MA Craig Hospital 3425 S. Clarkson Street Englewood, Colorado 80110 Dear Ms. Charlifue: HEALTH ONE ALLIANCE IJ'ISTilUTlONAUUVIE"( IIOARD As Chairman of the HealthONE Alliance Institutional Review Board I am pleased to that the revised informed consent for your aPProved protocol : "PartDers of Men witb Spinal Cord IDJury: Issues Related to Providing Assistance" was granted an expedited approval on August 9, 2001 by the HealthONE Alliance IRB subcommittee. Ths item wiD appear on lhe August 15, 200IIRB meeting agenda as information for the Board. Very truly yours; Jeffrey V. Matous, MD Chairman HealthONE Alliance lnstitutiolllll Review Board 1850 HiBh Stttet/ Denvn. Colorado 80l18/ Ph 303 839 .68) I I Fax 303 .U9.6836/ www. healthl.org 178

PAGE 194

HEALTHONE ALLIANCE August 30, 2000 Susan Cbarlifue Craig Hospital 3425 S. Clarkson Street Colorado 80110 Dear Ms. Cbarlifue: INSTITIJTIONAl'IUVIEW BOARD As Chairman of the HealthONE Alliance Institutional Review Board I am pleased to inform you that your submined research project and informed consent for: "Partners of Men with Spinal Cord InJury: Issues Related to Providing Assistance" have been granted anexpedited approval on August 30, 2000 by the HeallhONE Alliance IRB s\lbcomminee This protocol and consent form approval will appear on lhe next JRB meeting agenda as informational itelll!i for lhe Boanl members. This study will be due for aruJUaJ review in August 200 I You are approved as lhe Principal Investigator. David WeitzenJcamp, MS and Gale Whitened,, PhD are approved as Co-Investigators. The HealthONE Alliance IRB requires an annual repon on all approved projects, or if completed in less than one year, a final repon at lhe termination of the project. It is your responsibility after one year to eilher request reapproval or to advise the Board of study termination. In addition, it is essential that all significant changes in protocol, complications or unanticipated outcomes be reponed to this Board promptly Very truly c:!#t .. MD Chairman HealthONE Alliance Institutional Review Board J8SO High SITtet I Denver, Colorado BOJ\8/ Ph. 303.839.68) 1/ fax 303 .839.6836/......W.heahb\.oog 179

PAGE 195

REFERENCES Ainsworth, M.D. (1989). Attachments beyond infancy. Am Psycho), 44(4), 709-16. Ainsworth, M.D. S. (1969). Object relations, dependency, and attachment: A theoretical review of the infant -mother relationship. Child Development, 40( 4 ), 9691025. Alspaugh, M. E., Stephens, M. A., Townsend A. L., Zarit, S. H., & Greene, R. (1999). Longitudinal patterns of risk for depression in dementia caregivers: Objective and subjective primary stress as predictors. Psychol Aging, 14(1), 34-43. Alzheimer's Association. (2003). Statistics about Alzheimer's Disease [Web Page]. Retrieved October 16, 2003 from http://www alz.org/AboutAD/Statistics.htm. American Spinal Injury Association. (2002). American Spinal Injury Association: International Standards for Neurological Classification of Spinal Cord Injury. Chicago IL: American Spinal Injury Association. Arai, Y., Masui, K., Sugiura, M., & Washio, M. (2002): Fewer hours of care yet undiminished caregiver burden with new long-term care insurance in Japan. lnt J Geriatr Psychiatry, 17(5), 489-91. Arai, Y., & Ueda, T. (2003). Paradox revisited: Still no direct cmmection between hours of care and caregiver burden. lnt J Geriatr Psychiatry, 18(2), 188-9. Atkinson, R. (1998). The Life Story Interview (Qualitative Research Methods Series). Thousand Oaks CA: Sage. Barusch, A. S., & Spaid, W. M. (1989). Gender differences in caregiving: Why do wives report greater burden? Gerontologist, 29(5), 667-76. Bauer, J. J., & Bonanno, G. A. (2001). 1 can, I do, I am: The narrative differentiation of self-efficacy and other self-evaluations while adapting to bereavement. Journal of Research in Personality. 35, 424-448. Beach, S. R., Schulz, R., Yee, J. L., & Jackson, S. (2000). Negative and positive health effects of caring for a disabled spouse : Longitudinal findings from the caregiver health effects study. Psychol Aging. 15(2), 259-71. 180

PAGE 196

Beeson, R., Borton-Deutsch, S., Farran, C., & Neundorfer, M. (2000). Loneliness and depression in caregivers of persons with Alzheimer's disease or related disorders. Issues Ment Health Nurs, 21(8), 779-806. Beeson, R. A. (2003). Loneliness and depression in spousal caregivers ofthose with Alzheimer's disease versus non-caregiving spouses. Arch Psychiatr Nurs. 17(3), 13543. Bowlby, J. (1977). The making and breaking ofaffectional bonds: I. Aetiology and psychopathology in the light of attachment theory. British Journal of Psychiatry, 130, 201-210. Bowlby, J. (1980). Attachment and Loss, Volume 3: Loss (2nd ed.). New York: Basic Books. Bowlby, J. (1988). A Secure Base: Clinical Applications of Attachment Theory. London: Routledge. Boykin, A., & Winland-Brown, J. (1995). The dark side of caring: Challenges of caregiving. J Gerontol Nurs, 21(5), 13-8. Brody, E. M. (1981). "Women in the middle" and family help to older people. Gerontologist, 21(5), 471-80. Burton, L. C., Newsom, J. T., Schulz, R., Hirsch, C. H., & German, P. S. (1997). Preventive health behaviors among spousal caregivers. Preventive Medicine, 26, 162-169. Bush, H. A., & Job, S. A. (1993). Stressors of providing care to the elderly. AORN JournaL 57(4), 940-946. Cairney, J., Boyle, M., Offord, D. R., & Racine, Y (2003). Stress, social support and depression in single and married mothers. Soc Psychiatry Psychiatr Epidemiol, 38(8), 442-9. Chan, R. C. K., Lee, P. W. H., & Lieh-Mak, F. (2000). Coping with spinal cord injury: Personal and marital adjustment in the Hong Kong Chinese setting. Spinal Cord. 38(11), 687-696. Charlifue S., & Lammertse D.P. (2002). Aging in spinal cord injury. InS. Kirshblum, D.l. Campagnolo, & J.A. DeLisa (Eds.), Spinal Cord Medicine (pp. 409423). Philadelphia PA: Lippincott, Williams & Wilkins. 181

PAGE 197

Clark, M. C. & Standard, P. L. ( 1997). The caregiving story: How the narrative approach informs care giving burden. Issues in Mental Health Nursing, 18(2), 87-97. Cohen, S., Kamarck,T., & Mermelstein, R. (1983). A global measure of perceived stress. J Health Soc Behav, 24(4), 385-96. Collins, N. L., & Read, S. J. (1990). Adult attachment, working models, and relationship quality in dating couples. Journal of Personality and Social Psychology, 58(4), 644-663. Consortium for Spinal Cord Medicine. (1999). Outcomes following traumatic spinal cord injury: Clinical practice guidelines for health-care professionals. Washington DC: Paralyzed Veterans of America. Crewe, N. M., Athelstan, G. T., & Krumberger, J. (1979). Spinal cord injury: A comparison ofpreinjury and postinjury marriages. Arch Phys Med Rehabil, 60(6), 252-6. Crewe, N. M., & Krause, J. S. (1992). Marital status and adjustment to spinal cord injury. J Am Paraplegia Soc, 15(1 ), 14-8. Decker, S.D., Schultz, R., & Wood, D. (1989). Determinants ofwell-being in primary caregivers of spinal cord injured persons. Rehabilitation Nursing, 14(1 ), 6-8. DeVivo, M. J., & Fine, P.R. (1985). Spinal cord injury: Its short-term impact on marital status. Arch Phys Med Rehabil, 66(8), 501-4. Dijkers, M. P., Buda Abela, M., Gans, B. M., & Gordon, W. A. (1995). The aftermath of spinal cord injury. In: S.L. Stover, J.A. DeLisa, & G.G. Whiteneck (Eds.), Spinal Cord Injury: Clinical Outcomes from the Model Systems (pp. 185212). Gaithersburg MD: Aspen Publishing. Ditunno, J. F., Cohen, M. E., Formal, C., & Whiteneck, G. G. (1995). Functional Outcomes. In: S.L. Stover, J.A. DeLisa, & G.G. Whiteneck (Eds.), Spinal Cord Injury: Clinical Outcomes from the Model Systems (pp. 170-1184). Gaithersburg MD: Aspen Publishers. Douglas, S. L., & Daly, B. J. (2003). Caregivers of long-term ventilator patients: Physical and psychological outcomes. Chest, 123(4), 1073-81. 182

PAGE 198

Dowler, R., Richards, J. S., Putzke, J.D., Gordon, W., & Tate, D. (2001). Impact of demographic and medical factors on satisfaction with life after spinal cord injury: A normative study. J Spinal Cord Med, 24(2), 87-91. Duggan, C. H., & Dijkers, M. (1999). Quality of life peaks and valleys: A qualitative analysis ofthe narratives of persons with spinal cord injuries. Canadian Journal ofRehabilitation, 12(3), 179-189. Eastwood, E. A., Hagglund, K. J., Ragnarsson, K. T., Gordon, W. A., & Marino, R. J. (1999). Medical rehabilitation length of stay and outcomes for persons with traumatic spinal cord injury--1990-1997. Arch Phys Med Rehabil, 80(11 ), 1457-63. Elliott, T. R., Shewchuk, R. M., & Richards, J. S. (1999). Caregiver social problem solving abilities and family member adjustment to recent-onset physical disability. Rehabilitation Psychology, 44( 1 ), 1 04-123. eMedicine Consumer Health. (2003) Alzheimer Disease [Web Page]. Retrieved October 16, 2003 from http://www.emedicinehealth.com/ Articles/39500-1.asp. Evans, R. L. Bishop, D. S., & Ousley, R. T. (1992). Providing care to persons with physical disability: Effect on family caregivers. American Journal of Physical Medicine and Rehabilitation, 71(3), 140-144. Fenell, D. L. (1993). Characteristics oflong-term first marriages. Journal ofMental Health Counseling, 15(4), 446-460. Fitting, M., Rabins, P., Lucas, M. J., & Eastham, J. (1986). Caregivers for dementia patients: A comparison ofhusbands and wives. Gerontologist, 26(3), 248-52 Erankel, H. L., Coll, J. R., Charlifue, S. W., Whiteneck, G. G., Gardner, B. P., Jamous, M.A., Krishnan, K. R., Nuseibeh, I., Savic, G., & Sett, P. (1998). Long term survival in spinal cord injury: A fifty year investigation. Spinal Cord, 36(4), 266-74. Garland, D. E., Stewart, C. A., Adkins, R. H., Hu, S. S., Rosen, C., Liotta, F. J., & Weinstein, D. A. ( 1992). Osteoporosis after spinal cord injury. J Orthop Res, 1 0(3 ), 371-8. Gaynor, S. E. (1990). The long haul: The effects ofhome care on caregivers. IMAGE: Journal ofNursing Scholarship, 22(4), 208-212. 183

PAGE 199

Gerhart, K. A., Bergstrom, E., Charlifue, S. W., Menter, R. R., & Whiteneck, G. G. (1993 ). Long-term spinal cord injury: Functional changes over time. Arch Phys Med Rehabil, 74(1 0), 1030-4. Given, C. W., Collins, C. E., & Given, B. A. (1988). Sources of stress among families caring for relatives with Alzheimer's disease. Nurs Clin North Am, 23(1), 69-82. Glaser, R., Sheridan, J., Malarkey, W. B., MacCallum, R. C., & Kiecolt-Glaser, J. K. (2000); Chronic stress modulates the immune response to a pneumococcal pneumonia vaccine. Psychosom Med, 62( 6), 804-7. Goode, K. T., Haley, W. E., Roth, D. L., & Ford, G. R. (1998). Predicting longitudinal changes in caregiver physical and mental health: A stress process model. Health Psychology, 17(2), 190-198. Grant, G., Ramcharan, P., McGrath, M., Nolan, M., & Keady, J. (1998). Rewards and gratifications among family caregivers: Towards a refined model of caring and coping. Journal oflntellectual Disability Research, 42(Part 1 ), 58-71. Grant, I., Adler, K. A., Patterson, T. L., Dimsdale, J. E., Ziegler, M.G., & Irwin, M. R. (2002). Health cons.equences of Alzheimer's caregiving transitions: Effects of placement and bereavement. Psychosom Med, 64(3), 477-86. Guttman, R., Altman, R. D., & Nielsen, N.H. (1999). Alzheimer disease. Report of the Council on Scientific Affairs. Arch Fam Med, 8(4), 347-53. Hazan, C., & Shaver, P. (1987). Romantic love conceptualized as an attachment process. Journal ofPersonality and Social Psychology, 52(3), 511-524. Hazan, C., & Shaver, P.R. (1994). Authors' response: Deeper into attachment theory. Psychological Inquiry, 5, 68-79. Hedblad, B., Jonsson, S., Nilsson, P., Engstrom, G., Berglund, G., & Janzon, L. (2002). Obesity and myocardial infarction--vulnerability related to occupational level and marital status. A 23-year follow-up of an urban male Swedish population. I Intern Med, 252(6), 542-50. Hokby, A., Reimers, A., & Laflamme, L. (2003). Hip fractures among older people: Do marital status and type of residence matter? Public Health, 117 (3 ), 196-20 1. 184

PAGE 200

Holicky, R. (1996). Caring for the caregivers: The hidden victims of illness and disability. RehabilNurs, 21(5), 247-52. Holicky, R., & Charlifue, S. (1999). Ageing with spinal cord injury: The impact of spousal support. Disability and Rehabilitation, 21(5-6), 250-257. Johnson, N.J., Backlund, E., Sorlie, P. D., & Loveless, C. A. (2000). Marital status and mortality: The national longitudinal mortality study. Ann Epidemiol. 1 0( 4), 22438 Kiecolt-Glaser, J. K., Dura, J. R., Speicher, C. E., Trask, 0. J., & Glaser, R. (1991). Spousal caregivers of dementia victims: Longitudinal changes in immunity and health. Psychosom Med, 53( 4), 345-62. Kiecolt-Glaser, J. K., Glaser, R., Gravenstein, S., Malarkey, W. B., & Sheridan; J. (1996). Chronic stress alters the immune response to influenza virus vaccine in older adults. Proc Natl Acad Sci US A, 93(7), 3043-7. Kiecolt-Glaser, J. K., Marucha, P. T., Malarkey, W. B., Mercado, A.M., & Glaser, R. (1995). Slowing of wound healing by psychological stress. Lancet, 346(8984 ), 1194-6. Kiecolt-Glaser, J. K., Preacher, K. J., MacCallum, R. C., Atkinson, C., Malarkey, W. B., & Glaser, R. (2003). Chronic stress and age-related increases in the proinflammatory cytokine IL-6. PNAS, 1 00(15), 9090-9095. King, G., Cathers, T., Brown, E., Specht, J. A., Willoughby C., Polgar, J. M., MacKinnon, E., Smith, L. K., & Havens L. (2003). Turning points and protective processes in the lives of people with chronic disabilities. Qualitative Health Research, 13(2), 184-206. Knight, R. G., Williams, S., McGee, R., & Olaman, S. (1998). Caregiving and well being in a sample of women in midlife. Australian and New Zealand Journal of Public Health, 22(5), 616-620. Kobak, R. R., & Hazan, C. (1991). Attachment in marriage: Effects of security and accuracy of working models. Journal of Personality and Social Psychology. 60(6), 861-869. Kravdal, 0. (2001). The impact of marital status on cancer survival. Soc Sci Med, 52(3), 357-68. 185

PAGE 201

Kreuter, M. (2000). Spinal cord injury and partner relationships. Spinal Cord, 38(1), 2-6. Kreuter, M., Sullivan, M., Dahllof, A. G., & Siosteen, A. (1998). Partner relationships, functioning, mood and global quality of life in persons with spinal cord injury and traumatic brain injury. Spinal Cord, 36(4), 252-61. Lazarus, R. S., & Folkman, S. (1984a). Stress. Appraisal and Coping. New York: Springer. Lazarus, R. S., & Folkman, S. (1984b). Coping and adaptation. In W.D. Gentry (Ed.), Handbook ofBehavioral Medicine. New York: Guilford. Leblanc, A. J., London, A. S., & Aneshensel, C. S. (1997). The physical costs of AIDS caregiving. Soc Sci Med, 45(6), 915-23. Lee, S., Colditz, G. A., Berkman, L. F., & Kawachi, I. (2003). Caregiving and risk of coronary heart disease in U.S. women: A prospective study. Am J Prev Med, 24(2), 113-9. Lin, C. C., Rogot, E., Johnson, N.J., Sorlie, P. D., & Arias, E. (2003). A further study of life expectancy by socioeconomic factors in the National Longitudinal Mmtality Study. Ethn Dis, 13(2), 240-7. Mahoney, A., Pargament, K. 1., Jewell, T., Swank, A. B. S. E., Emery, E., & Rye, M. (1999). Marriage and the spiritual realm: The role of proximal and distal religious constructs in marital functioning. Journal of Family Psychology. 13(3), 321-338. Marchi-Jones, S., Murphy, J. F., & Rousseau, P. (1996). Caring for the caregivers. Journal of Gerontological Nursing, 22(8), 7-13. Marks, S. R. (1977). Multiple roles and role strain: Some notes on human energy, time, and commitment. American Sociological Review, 42, 921-936. McAdams, D. (1995). The Life Story Interview (Revised). Chicago: Northwestern University. McAdams, D.P. (1990). The Life Story Interview. Chicago: Northwestern University. McAdams, D.P. (1993). The Stories We Live By: Personal Myths and the Making of the Self. New York: Guilford. 186

PAGE 202

McAdams, D.P. (2001). Coding Autobiographical Episodes for Themes of Agency and Communion. Northwestern University. Retrieved November 22, 2003 from http:/ /www.sesp.northwestetn.edu/foley/instruments/ Agency_ CommunionO 1.pdf. McAdams, D.P., Diamond, A., de St. Aubin, E., & Mansfield, E. (1997). Stories of commitment: The psychosocial construction of generative lives. Journal of Personality & Social Psychology, 72(3), 678-694. Menter, R. R. (1993). Spinal cord injury and aging: Exploring the unknown. 1993 Heiner Sell Lecture of the American Spinal Injury Association. J Am Paraplegia Soc, .1(4), 179-89. Mikulincer, M., & Florian, V. (1999). The association between spouses' self-reports of attachment styles and representations offamily dynamics. Family Process, 38(1), 69-83. Miller, B., & Cafasso, L. (1992). Gender differences in caregiving: Fact or artifact? Gerontologist, 32(4), 498-507. Miller, B., Campbell, R. T., Farran, C. J., Kaufman, J. E., & Davis, L. (1995). Race, control, mastery, and caregiver distress. Journal of Gerontology, 50B(6), S374-S382. Mintz, S. (1996). Caregiver disorder: An unacknowledged illness. Take Care, 5, 110. Mirowsky, J., & Ross, C. E. (1990). Control or defense? Depression and the sense of control over good and bad outcomes. J Health Soc Behav, 31(1), 71-86. National Spinal Cord Injury Statistical Center. (2003). Annual Report for the Model Spinal Cord Injury Care Systems. Birmingham AL: National Spinal Cord Injury Statistical Center. Navaie-Waliser, M., Spriggs, A., & Feldman, P. H. (2002). Informal caregiving: Differential experiences by gender. Med Care, 40(12), 1249-59. Neundorfer, M. M., McClendon, M. J., Smyth, K. A., Stuckey, J. C., Strauss, M. E., & Patterson, M. B. (200 1 ). A longitudinal study of the relationship between levels of depression among persons with Alzheimer's disease and levels of depression among their family caregivers. J Gerontol B Psychol Sci Soc Sci, 56(5), P301-13. Nosek M.A. (1990). Personal assistance: Key to employability of persons with physical disabilities. Journal of Applied Rehabilitation Counseling, 21, 3-8. 187

PAGE 203

Novak, M., & Guest, C. (1989). Application of a multidimensional caregiver burden inventory. The Gerontologist, 29(6), 798-803. O'Brien, M. T. (1993). Multiple sclerosis: Stressors and coping strategies in spousal caregivers. Journal of Community Health Nursing, 10(3), 123-135. Ochberg, R. L. (1988). Life stories and the psychosocial construction of careers. Journal ofPersonality, 56(1), 173-179; 202-203. Odell, M., & Quinn, W. H. (1998). Congruence, desire for change, and adjustment during the first year of marriage. Marriage & Family Review, 27(1/2), 91-112. Parris Stephens, M. A., Franks, M. M., & Townsend, A. L. (1994). Stress and rewards in women's multiple roles: The case of women in the middle. Psychology and Aging, 9(1), 45-52. Patton, M. Q. (2002a). Designing Qualitative Studies. In M. Q. Patton (Ed.), Qualitative Research and Evaluation Methods (3 ed., pp. 241-242). Thousand Oaks, CA: Sage Publications. Patton, M. Q. (2002b). Qualitative Analysis and Interpretation. In M. Q. (Ed.) Qualitative Research and Evaluation Methods (3rd edition ed., p. 453). Thousand Oaks, CA: Sage Publications. Pearlin, L. I. (1989). The sociological study of stress. J Health Soc Behav, 30(3), 241-56. Pearlin, L. 1., Mullan, J. T., Semple, S. J., & Skaff, M. M. (1990). Caregiving and the stress process: An overview of concepts and their measures. The Gerontologist, 30(5), 583-594. Plechaty, M. ( 1987). Perceptual congruence of five attitudes among satisfied and unsatisfied couples. Psychological Reports, 61 (2), 527-53 7. Post, M. W., de Witte, L. P., van Asbeck, F. W., van Dijk, A. J., & Schrijvers, A. J. (1998). Predictors ofhealth status and life satisfaction in spinal cord injury. Arch Phys Med Rehabil, 79( 4 ), 395-401. Radloff, L. S. (1977). The CES-D Scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 1, 3 85-401. 188

PAGE 205

Stein, H., Jacobs, N.J., Ferguson, K. S., AlJen, J. G., & Fonagy, P. (1998). What do adult attachment scales measure? Bulletin of the Menninger Clinic, 62(1 ), 33-82. D. E., K., & Kercher, K. (I 994). Caregiver burden and generic wellbemg: Opposite sides ofthe same coin? Gerontologist, 34(1), 88-94. Tedeschi, R. G., & Calhoun, L. G. (1995a). Support and intervention. In R. G. Tedeschi, & L. G. Calhoun (Eds.), Trauma and Transformation: Growing in the Aftermath of Suffering. Thousand Oaks, CA: Sage. Tedeschi, R. G., & Calhoun, L. G. (1995b). Tasks and cognitive processes in coping with trauma. In R. G. Tedeschi, & L. G. Calhoun (Eds.), Trauma and Transformation: Growing in the Aftermath of Suffering Thousand Oaks, CA: Sage. Tower, R. B., & Kasl, S. V. (1995). Depressive symptoms across older spouses and the moderating effect of marital closeness. Psycho} Aging, 10(4), 625-38. Uniform Data System for Medical Rehabilitation. Guide for the Uniform Data Set for Medical Rehabilitation (including the FIMSM instrument) (Version 5.0 ed.). Buffalo NY: State University ofNew York. Vedhara, K., Cox, N. K., Wilcock, G. K., Perks, P., Hunt, M., Anderson, S., Lightman, S. L., & Shanks, N. M. (1999). Chronic stress carers of ,., dementia patients and antibody response to influenza vaccmatwn. Lancet, 353(915_,), 627-31. 1 J M Zhang J. Savage, M. V., Hirsch, I. B., & Siegler, I. Vitaliano, P. of st;es;, the metabolic syndrome, and coronary C. (ZOOZ). A pa M d 64(3) 418-35. .. Psychosom e heart dlsease. -which to explore close EM. (2000). A secure base from "C & : ent. 71 (l)' 164-172. \:J.,c n m H L Williams, D., & n A. h. s R ben ., d presslO relations lp "M Kleber, H., u and secondary e. nd 1>. A vottencser, :Offi patternS }n ddictS, a\cohohCS, a N\ d op1ate a \). \_\ 911). XleS witb 854-62. G., & \'\\\\\\'0.\1 \\fbiteneck, G. t \W \ S nwacto j be aO.aed 1o"1. S / f : \ \ ''\ \ 1 ':. \ \

PAGE 206

Radloff, L. S., & Teri, L. (1986). Use of the Center for Epidemiological Studies Depression Scale with older adults. Clinical Gerontologist, 5(1 /2), 119-136. Richard, L. S., Wakefield Jr., J. A., & Lewak, R. (1990). Similarity of personality variables as predictors of marital satisfaction: A Minnesota Multiphasic Personality Inventory (MMPI) item analysis. Personality and Individual Differences. 11 (1 ), 3943. Roberts, R. E., & Vernon, S. W. (1983). The Center for Epidemiologic Studies Depression Scale: Its use in a community sample. Am J Psychiatry, 140(1 ), 41-6. Sarafino, E. P. (1990) Stress, illness and coping In E.P. Sarafino (Ed.), Health Psychology: Biopsychosocial Interactions New York: John Wiley & Sons. Schulz, R., & Beach, S. R. (1999). Caregiving as a risk factor for mortality: The Caregiver Health Effects Study. JAMA, 282(23), 2215-2219. Schulz, R., & Decker, S. (1985). Long-tern1 adjustment to physical disability: The role of social support, perceived control, and self-blame. J Pers Soc Psycho}, 48(5), 1162-72. Schulz, R., Tompkins, C. A., & Rau, M. T. (1988). A longitudinal study ofthe psychosocial impact of stroke on primary support persons. Psycho! Aging, 3(2), 13141. Schulz, R., Visintainer, P., & Williamson, G. M. (1990). Psychiatric and physical morbidity effects of caregiving. Journal of Gerontology, 45(5) 181-191. Shaw, W S., Patterson T. L., Ziegler, M.G., Dimsdale J. E. Semple, S. J., & Grant, I. (1999). Accelerated risk of hypertensive blood pressure recordings among Alzheimer caregivers J Psychosom Res. 46(3), 215-27. Shewchuk, R. M., Richards J. S., & Elliott, T. R. (1998). Dynamic processes in health outcomes among caregivers of patients with spinal cord injuries Health Psychology. 17(2), 125-129. Simpson, J. A., & Rholes, W. S. (1997). Attachment in adulthood. In: J. A. Simpson & W. S. Rholes (Eds.), Attachment Theory and Close Relationships (pp. 3-21 ). New York: Guilford Press. Smith, K. R., & Zick C. D. (1994). Linked lives, dependent demise? Survival analysis of husbands and wives. Demography. 31 (1 ), 81-93. 189

PAGE 207

Stein, H., Jacobs, N.J., Ferguson, K. S., Allen, J. G., & Fonagy, P. (1998). What do adult attachment scales measure? Bulletin of the Menninger Clinic, 62(1), 33-82. Stull, D. E., Kosloski, K., & Kercher, K. (1994). Caregiver burden and generic wellbeing: Opposite sides ofthe same coin? Gerontologist, 34(1), 88-94. Tedeschi, R. G., & Calhoun, L. G. (1995a). Support and intervention. In R. G. Tedeschi, & L. G. Calhoun (Eds.), Trauma and Transformation: Growing in the Aftermath of Suffering Thousand Oaks, CA: Sage. Tedeschi, R. G., & Calhoun, L. G. (1995b). Tasks and cognitive processes in coping with trauma. In R. G. Tedeschi, & L. G. Calhoun (Eds.), Trauma and Transformation: Growing in the Aftermath of Suffering Thousand Oaks, CA: Sage Tower, R. B., & Kasl, S. V. (1995). Depressive symptoms across older spouses and the moderating effect of marital closeness. Psychol Aging, 10(4), 625-38. Uniform Data System for Medical Rehabilitation. Guide for the Uniform Data Set for Medical Rehabilitation (including the FIMSM instrument) (Version 5.0 ed.). Buffalo NY: State University of New York. Vedhara, K., Cox, N. K., Wilcock, G. K.,Perks, P., Hunt, M., Anderson, S., Lightman, S. L., & Shanks, N. M. (1999): Chronic stress in elderly carers of demehtia patients and antibody response to influenza vaccination. Lancet, 353(9153), 627-31. Vitaliano, P. P., Scanlan, J. M., Zhang, J., Savage, M. V., Hirsch, I. B., & Siegler, I. C. (2002). A path model of chronic stress, the metabolic syndrome, and coronary heart disease. Psychosom Med, 64(3), 418-35. waters, E., & Cummings, E. M. (2000). A secure base from which to explore close relationships. Child Development, 71 (1 ), 164-172. Weissman, M. M., Pottenger M.,K.Jeber H ., Ruben, H. L., Williams D., & Thompson, W. D. (1977). Symptom patterns in primary and secondary depression. A comparison of primary depressives with depressed opiate addicts, alcoholics, and schizophrenics. Archives of General Psychiatry. 34(7), 854-62. Weitzenkamp, D. A., K. A., Charlifue, S. W., Whiteneck, G. G., & Savic, G. (1997). Spouses of spinal cord injury survivors: The added impact of caregiving. Archives of Physical Medicine and Rehabilitation, 78(8), 822-827. ,. 190 i -y / ;:.

PAGE 208

Williamson, G. M. Shaffer D. R., & Schulz, R. (1998). Activity restriction and prior relationship history as contributors to mental health outcomes among middle-aged and older spousal caregivers. Health Psychology, 17(2), 152-162. Wolfson, C. Wolfson, D. B., Asgharian M., M'Lan, C. E., Ostbye, T., Rockwood, K., & Hogan, D. B. (2001). A reevaluation ofthe duration of survival after the onset of dementia. N Engl J Med, 344(15) 1111-6. Zarit S. H. Reever, K. E. & Bach-Peterson, J. (1980). Relatives ofthe impaired elderly: Correlates offeelings ofburden. Gerontologist, 20(6) 649-55. Zarit, S. H., Todd, P. A., & Zarit, J. M. (1986). Subjective burden ofhusbands and wives as caregivers: A longitudinal study. The Gerontologist, 26(3), 260-266. ,' 191