Psychological factors and quality of life


Material Information

Psychological factors and quality of life a prospective study of cardiac transplantation recipients
Alternate title:
Prospective study of cardiac transplantation recipients
Physical Description:
ix, 123 leaves : ill. ; 29 cm.
Sears, Samuel F., 1968-
Publication Date:


Subjects / Keywords:
Research   ( mesh )
Quality of Life   ( mesh )
Heart Transplantation -- psychology   ( mesh )
Psychological Tests   ( mesh )
Reproducibility of Results   ( mesh )
Predictive Value of Tests   ( mesh )
Immunosuppression   ( mesh )
Spouses   ( mesh )
Department of Clinical and Health Psychology thesis Ph.D   ( mesh )
Dissertations, Academic -- College of Health Related Professions -- Department of Clinical and Health Psychology -- UF   ( mesh )
bibliography   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph.D.)--University of Florida, 1995.
Bibliography: leaves 106-113.
Statement of Responsibility:
by Samuel F. Sears, Jr.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002306438
oclc - 50281628
notis - ALQ9756
System ID:

This item is only available as the following downloads:

Full Text








This paper is dedicated to the love, support, and patience of my entire

family. I wish to thank my parents, Mr. and Mrs. Samuel F. Sears, my

grandparents, Mr. and Mrs. L. M. Willis, my brother and sister-in-law, Mr.

and Mrs. Christian D. Sears, and my mother and father-in-law, Mr. and Mrs.

Charles R. Evans. My family provided both practical and emotional support

during the challenges of maturing into a competent psychologist. In

addition, my wife, Staci, deserves credit for her belief and faith in me. Her

powerful love and patience has allowed me to reach my highest goals.

My close friend, Russell G. Hoffmann III, deserves special mention for

his strong loyalty and friendship during our graduate training together. Our

mutual support and respect has helped to shape me into a better person and

a well-rounded professional.

A substantial amount of time and effort necessary for the realization

of this project was supplied by my research team members. Kristin DiJoris,

Tore Gintolli, Tanya Hiller, and Daniel Sagastizabal enthusiastically

conducted numerous telephone interviews in a highly professional manner.

The cardiac transplantation team was another group of people that greatly

assisted me in this work. I would like to thank the transplant coordinators,

Chuck McGinn, R.N., Susan Selman, R.N, and Tracy Walker, R.N., and the

transplant computer manager, John Posser, and the transplant services

secretary, Denise Bishop. In addition, I am indebted to Roger M. Mills Jr.,

M.D., the medical director of the cardiac transplantation team, for his letter

of support for this project.

Finally, I am greatly appreciative of the work of my doctoral

committee members, Dr. Hugh Davis, Dr. Eileen B. Fennell, Dr. John Newell,

and Dr. Anthony F. Greene. I have great respect and admiration for each of

these individuals and am honored for their participation in this study. Finally,

I wish to provide a special thanks to Dr. James R. Rodrigue for serving as

my mentor and providing enormous support for my continuing development.


ACKNOW LEDGMENTS ................................... ii

LIST OF TABLES ................ .. .................... vi

LIST OF FIGURES ................ ................... vii

ABSTRACT ...............................................viii

INTRODUCTION ........................................1

REVIEW OF LITERATURE .................................. 3

Stages of Transplantation ............................. 6

Diagnosis and Decision Making ..................... 6
Proceeding with the Candidate Selection Process ........ 7
The W waiting Period ............................ 9
Transplantation and Post-Hospitalization ............. 10
Long-Term Rehabilitation and Recovery .............. 11

Quality of Life .................................... 12

Conceptualization ............................. 12
Measurement Approaches ............. ......... 13
Predictors of Quality of Life ................... 18
Quality of Life by Proxy ......................... 24

Quality of Life in Cardiac Transplantation ................. 29

Review of Studies .............................29
Summary and Conclusions ....................... 39

Previous Research at the University of Florida .............. 41

Overview ...................................... 45

Hypotheses .................................... 46

METHOD ................ .......................... 49

Subjects........................................... 49
Procedures ...................................... 51
M measures ........................................52
Post-transplantation Assessment Battery ... .............. 57

RESULTS .................. ........................ 61

DISCUSSION .........................................93

REFERENCES .........................................106

APPENDIX ..........................................114

BIOGRAPHICAL SKETCH ................................. 123

Table Page

1 Quality of Life Measures ............................. 16

2 Demographic Data .................................50

3 Pre-transplantation Intellectual and Cognitive Functioning ...... 62

4 MMPI Means of Candidates and Recipients ................ 65

5 Quality of Life Grand Means ..........................76

6 Mean Recipient Quality of Life Ratings By Cluster Group ....... 85

7 Mean Spouse/Proxy Quality of Life Ratings By Cluster Group .... 88

8 Quality of Life Means for Planned Comparisons ............. 92


1 Symptom Frequencies: Recipient Report ............

2 Symptom Frequencies: Spouse/Proxy Report .........

3 Problem Totals: Recipient Report .................

4 Problem Totals: Spouse/Proxy Report ..............

5 Cardiac Transplantation Candidate MMPI Cluster Groups:
Four Group Solution ..........................

6 Cardiac Transplantation Candidate MMPI Cluster Groups:
Three Group Solution .........................


...... 67

...... 69

...... 71

...... 73



Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



Samuel F. Sears, Jr.

August 1995

Chairperson: Hugh Davis, Ph.D.
Cochairperson: James R. Rodrigue, Ph.D.
Major Department: Clinical and Health Psychology

The purpose of the present study was four-fold: 1) to provide

descriptive information about the prevalence of immunosuppressive

symptomatology and quality of life of cardiac transplantation recipients, 2)

to test the predictive validity of a pre-transplantation psychological

assessment battery on the recipient's ratings of post-transplantation quality

of life, 3) to test the predictive validity of the pre-transplantation

psychological assessment battery on the spouse/proxy ratings of post-

transplantation quality of life, and 4) to assess the correspondence of the

patient and proxy raters on quality of life.

Results indicated that on average, cardiac transplantation recipients

commonly experience approximately 10 immunosuppressive symptoms;

fatigue and mood swings were noted as particularly problematic by the

recipients and spouse/proxies. Following the formation of four cluster

groups of pre-transplantation MMPI profiles, tests of significance found no

differences in the quality of life variables between cluster groups.

Regression analyses were also used to test the predictive validity of the

other pre-transplantation medical and psychological variables. Results from

these analyses found that trait anxiety was a significant predictor of

increased symptom frequency and symptom problems, and decreased

mental health. Similar results were found using the spouse/proxy reports,

such that trait anxiety was again a significant negative predictor of mental

health functioning and satisfaction with life. Taken together, trait anxiety

emerged as a diverse predictor of post-transplantation quality of life

functioning, as viewed from both the recipient and spouse/proxy

perspectives. Finally, results from correlational analyses between recipient

and spouse/proxy reports yielded generally high correspondence, but planned

comparisons found significant differences in mental health functioning, with

spouse/proxies rating the recipient's mental health lower. Collectively,

modest support for the use of pre-transplantation psychological variables as

predictors of post-transplantation quality of life was found.


Cardiac transplantation has rapidly emerged as a viable treatment

option over the past decade. The discovery of reliable immunosuppression

drugs that decrease organ rejection episodes has made cardiac

transplantation the treatment of choice for individuals diagnosed with end-

stage cardiac disease. The research on longevity in the vast majority of

recipients suggests that cardiac transplantation extends the length of life

(Mai, McKenzie, & Kostuk, 1990). Therefore, cardiac transplantation is

expected to become a widespread procedure in future health care that may

be limited primarily by the availability of donor organs. Because longevity of

cardiac transplantation recipients is now established, researchers and health

policy administrators have also begun to consider outcomes in terms of

quality of life post-transplantation. Although much disagreement exists

about the definition, measurement and theoretical basis of quality of life,

initial investigations of post-transplantation quality of life have agreed that

the recipients that survive the transplantation waiting and initial post-

operative periods have significant improved quality of life from pre-

transplantation to post-transplantation (Jones, Taylor, Downs, & Spratt,

1992; O'Brien, Banner, Gibson, & Yacoub, 1988). As a result of these

studies, Freeman, Folks, Sokol, and Fahs (1988) have suggested that "the

most important clinical research question concerns quality of life following

cardiac transplantation" and research should "continue to examine which

assessment and management techniques are beneficial to the cardiac

recipient in an attempt to identify clinical and demographic characteristics

that predict an enhanced quality of life" (pp. 53-54).

The current study was a longitudinal investigation of cardiac

transplantation recipients of Shands Hospital. Telephone interviews were

used to assess the self-perceived quality and satisfaction of life of each

recipient. Additional ratings of the recipients' quality and satisfaction of life

were collected by sampling the responses of the recipient's spouse or

significant other (proxy) on similar measures. Interview responses from the

cardiac transplantation recipients and their spouse/proxies were then merged

with information from the pre-transplantation medical and psychological

evaluations that are routinely performed as part of the candidate selection

process. This joining of information allowed for the examination of pre-

transplantation medical and psychological variables that were associated

with desired post-transplantation outcome in terms of quality of life.

Possible differences were examined between established personality clusters

of cardiac transplantation candidates (Robinson, Greene, & Geisser, 1993;

Hecker, Norvell, & Hills, 1989) in the dimensions of quality of life. Follow-

up regression analyses were performed to find variables that may serve as

useful "markers" of successful post-transplantation quality of life. Data from

this project will be very helpful to transplantation teams in planning

interventions with transplantation candidates in order to ensure the highest

quality of life possible for each patient at post-transplantation.


As recently as 15 years ago, the diagnosis of end-stage cardiac

disease provided the patient with few treatment options and imminent

death. Cardiac transplantation was considered an experimental procedure

that was rarely performed. Prior to 1980, the Registry of the International

Society for Heart Transplantation listed only 360 recipients worldwide (Kriett

& Kaye, 1990). The discovery of reliable immunosuppression drugs, such as

cyclosporine, sparked a dramatic increase in.transplantation centers from 8

in 1980 to 149 in 1989 and a dramatic increase in transplantation recipients

that eventually resulted in an organ shortage (Kriett & Kaye, 1990).

National statistics indicate that the most common diagnoses in cardiac

transplantation candidates are cardiomyopathy and coronary artery disease

(Kriett & Kaye, 1990). Cardiomyopathy refers to a set of disorders that

includes structural or functional abnormalities of the heart muscle (Mersch,

1985). Coronary artery disease is an atherosclerotic process that involves a

thickening of the intimal layer of the blood vessel wall which allows for the

accumulation of occluding lipids and manifests itself in four major clinical

syndromes: angina pectoris, myocardial infarction, heart failure, or sudden

death (Brannon, Foley, Starr, & Black, 1993). The family of

cardiomyopathies important to transplantation include ischemic, idiopathic,

viral, and dilated. Ischemic cardiomyopathy refers to the lack of oxygen

supply to cardiac tissue usually due to the presence of coronary artery

disease. Idiopathic cardiomyopathy refers to cardiac muscle cell death due

to unknown causes. Some researchers believe that some of the unknown

causes of idiopathic cardiomyopathy may be undetected viruses (Lorell &

Braunwald, 1992). Viral cardiomyopathy is the inflammation of the visceral

and parietal pericardial membranes due to the presence of many common

viruses that cause such diseases as mumps, chicken pox, mononucleosis,

and influenza. Dilated cardiomyopathy refers to the enlargement and

impaired systolic function of one or both ventricles (Wynne & Braunwald,

1992). The end-result of these cardiomyopathies often is congestive heart

failure which is characterized by the inability of the heart to maintain an

adequate cardiac output to meet the demands of tissue. The symptom

presentation may include shortness of breath, cardiac pain, and lethargy.

The ability of the heart to perform its work is captured in an expression

known as an ejection fraction. The ejection fraction is a global index of the

extent of ventricular fiber shortening and is thought to be a useful measure

of overall left ventricular pump function (Braunwald, 1992). The average

ejection fraction for adults is .67 t .08, while the ejection fraction for end-

stage cardiac disease patients is less than .20.

Because historically only palliative care existed for patients suffering

from end-stage heart failure, much excitement was generated by the

potential to become a transplantation center in the mid 1980's. The growth

of transplantation centers was managed by the National Organ

Transplantation Act of 1984 which required centers to maintain rigid

selection criteria and success rates to become transplantation centers. The

establishment of multi-disciplinary teams of health professionals has become

an important element to both the pre- and post-transplantation periods by

aiding in selection and follow-up care. The survival rates in the post-

transplantation cyclosporine era appear to be encouraging with 1 year

longevity ranging from 83-95% (Kriett & Kaye, 1990). Recent follow-up

data indicate that approximately 60-70% of patients are still alive after 5

years (Kaye, 1993).

A major area neglected by researchers in cardiac transplantation is the

consideration of individual differences. For example, the differential effects

of gender on the presentation, consideration, and receipt of cardiac

transplantation remains a generally unexplored area. Research on women

and general heart disease also appears limited and usually involves the

generalization of research findings from studies of men to women. Eaker

(1989) reviewed the psychosocial characteristics that have been studied in

women with heart disease and found little evidence for generalizing findings

from studies of men to women. For example, Eaker concluded that

constructs such as Type A behavior, a competitive, ambitious, and hard-

driving lifestyle, do not seem to be related to the incidence of heart disease

but may be related to chest pain symptomatology. Haynes and Czajkowski

(1993) drawing on more recent research drew tentative conclusions about

other psychosocial risk factors. These authors concluded that a strong

inverse association between education level and heart disease exists for

women. In addition, suppression of anger and lack of social support appear

to be somewhat linked to heart disease in women. More importantly, these

authors continued to emphasize the lack of sufficient research with women

with heart disease. The problems of insufficient sample sizes in studies with

women with heart disease are magnified in the cardiac transplantation

literature because of the already small potential samples of cardiac

transplantation recipients. It is suggested that as the prevalence of cardiac

transplantation progresses, research attention to the specific needs of

women in transplantation will be necessary to provide a more complete

understanding of the effectiveness of protocols for women.

States of Cardiac Transplantation

Clinical experience with the heart transplantation candidate over the

course of care has indicated that stages exist that require differing demands

of the patient. A number of researchers have provided overviews of the

stages of cardiac transplantation (Kuhn, Davis, & Lippman, 1988; Mischel &

Murdaugh, 1987). This section will summarize the stages of cardiac

transplantation as they relate to the subjects of the current study. It should

be noted that these stages represented anecdotal experiences of clinicians

that have worked in this area rather than stages that have been empirically


Diagnosis and Decision Making

The diagnosis of a life-threatening condition naturally elicits a variety

of reactions such as anxiety, mood disturbance, panic and fear. The

announcement of the diagnosis requires the person to accept that their

condition is terminal and, in most instances, confront their own mortality.

The amount of time available for this acceptance process is variable; some

patients have a period of months or even years to adapt, others have just a

few days. In a paper presented to the American Psychological Association

(1992), Greene reported that the average number of months of illness for

cardiac transplantation candidates was 77.16 (sd = 90.37) but the range

was 1-660 months. Some researchers have speculated that a relationship

between onset of cardiac difficulties and psychological distress exists such

that patients with rapid onset of cardiac disease (e.g. viral cardiomyopathy)

exhibit more psychological difficulties (Freeman et al., 1988). To be

considered a cardiac transplantation candidate, the patient must be classified

in class 4 (patient exhibits discomfort at rest) of the New York Heart

Association classification system. If the experience of cardiological

difficulties is long-term, the patient may have become resolved to a

premature death. Following the acceptance and/or consideration of the

severity of the diagnosis, the patient and his/her significant others are faced

with the decision to pursue transplantation. The transplantation alternative

is an object of approach-avoidance for the patient that often brings thoughts

and concerns of the negative side effects of the transplantation such as

rejection, continuous medical follow-up, and possibly death.

Proceeding with the Candidate Selection Process

After acknowledging the terminal diagnosis and deciding to proceed

with the transplantation program, the pre-transplantation assessment begins.

This stage's importance is underlined because it is the period in which all

possible relevant information should be obtained for the purpose of

preparation. Unlike some other areas of medicine which treat difficulties

once they arise, the transplantation program must prevent or foresee major

impediments to the procedure or to recovery. More clearly, trained health

professionals must make every attempt to identify the stressors and

resources of a patient and significant others. Many national transplantation

centers have indicated the use of a multi-disciplinary team for assessment

(Kay & Bienenfeld, 1991). The assessment of coping requires a

comprehensive review of the patient's environmental system as well as their

own personal resources. Anecdotally, it would appear that at the very same

time the stressors are at their peak, transplantation candidates' resources

are at their most diminished point. Supplementary resources must be

integrated into treatment to minimize complications and setbacks.

Therefore, the multi-disciplinary approach is useful to provide an assessment

of all possible resources and stressors for the patient.

The pre-transplantation evaluation represents another significant step

for the cardiac transplantation candidate. Patients are faced with the task of

invasive medical diagnostic procedures and repeated interviews. At Shands

Hospital, candidates and their families are interviewed by a psychological

team, a social worker, and a pastoral counselor. Each of these evaluations

include the disclosure of personal information. In addition, psychological

test batteries are performed to compare the candidate to a normative sample

of cardiac transplantation candidates. The patient's responses and

perseverance through these evaluations often serve as important examples

of the patient's potential future interactions with the transplantation team

members. These evaluations can sometimes be "in vivo" tests of the

patient's ability to comply with the medical regimen. The demand

characteristics of the evaluation for the patient may be perceived as

attempting to present yourself in the best possible light, so as to facilitate

the patient being "listed" on the organ availability list. Current general

contraindications to transplantation include over age 65, being diagnosed

with cancer, an active infection, Type II Diabetes, severe vascular disease,

renal dysfunction, or a psychosis (Braunwald, 1992).

The Waiting Period

Although the listing of the patient is certainly somewhat relieving, the

waiting period that follows may represent the most difficult time from the

patient's perspective. The potential for problems is great. During this stage,

many cardiac transplantation candidates experience increased death anxiety

which may manifest itself in insomnia due to concerns that the patient may

die in his sleep. Anxiety appears to increase in relation to the deteriorating

physical health. The patient's anxiety during this period may be warranted

because it is suggested that 20-30% of cardiac transplantation candidates

expire on the waiting list for available organs (Braunwald, 1992). Another

difficulty for transplantation candidates is disease-related dementia. In

absolute value terms, these conditions represent mental status decline, thus,

a further depletion of resources that are already taxed. Difficulties with

information comprehension and memory impairments in encoding and

retrieval are often reported. Although little can be done to prevent this

condition, the patient and family may be aided by the education components

that are a routine part of the previous stages. The patient and his family can

often recognize these symptoms instead of experiencing alienation and

anxiety about a medical decline. Other common problems for the transplant

candidate includes the 24 hour "on-call status" for notification as to the

possibility of transplantation. However, sometimes patients respond to false

alarms in which they are paged to report for surgery only to be informed that

the available organ was not acceptable for harvesting from a potential donor.

Transplantation and Post-hospitalization

Because much work has been performed to arrive at transplantation,

the actual transplantation surgery can be anti-climactic. The primary

difficulties with surgery are the post-surgical complications both medically

and psychologically. The surgery is usually performed on someone whose

condition was not very good and the waiting period often drains a patient of

his immediate resources. Medical difficulties include problems with

tolerating the surgery and the subsequent biopsies. There are currently no

non-invasive techniques for detecting rejection of the transplanted tissue.

Transplantation recipients must undergo the insertion of a bioptome through

their right jugular vein and small samples of tissue are excised (Reitz, 1992).

In addition to repeated biopsies, recipients must cope with the various side

effects of the administration of immunosuppressive drugs. Lough, Lindsey,

Shinn, and Stotts (1987) outlined the most common complications

associated with cyclosporine therapy including: renal dysfunction,

hypertension, interstitial myocardial fibrosis, and increased susceptibility to

infection. The existence of Cytomegalovirus (CMV) appears to be highly

prevalent (Braunwald, 1992). Psychologically, patients suffer from guilt

about the death of the donor and sometimes indebtedness to the medical

staff for saving their life. Nonetheless, both medical and psychological

difficulties represent threats that require coping.

Lona-Term Rehabilitation and Recovery

Numerous clinic visits which include hemodynamic measurements and

tissue biopsies are required during this period for the purpose of

assessing/monitoring immunosuppression levels and possible rejection in the

donor organ. Obviously, these visits can be filled with some anxiety about

the findings. Restarting life with a transplantation represents its own

difficulties. The adjustment for the patient and his family can be

monumental. New roles are established and necessitated. Attaining gainful

employment is a crucial concern for non-retired transplantation recipients.

Some transplantation centers consider return to work as a necessary

condition for transplantation (Paris, Woodbury, & Thompson et al., 1992).

Social and personal concerns are also highly prevalent. Other people see the

patient as a "transplant recipient" and the surgical mask over the patient's

face reminds them of their special status. Clearly, body image concerns are

also a significant issue for the patient as he/she considers the foreign body

parts and scars that are part of his/her new life (Lough et al., 1987). Quality

of life concerns became an essential focus of the post-transplant period.

Many patients are left with more questions and decisions to make than ever

before in their adult life. Once again, problem-solving is helpful in facilitating

the patient's adaptation to the new roles and may enhance adherence.

Quality of Life


The current emphasis on the consideration of quality of life may have

begun in 1964 when President Lyndon B. Johnson stated that the goal of

The Great Society was "to improve the quality of life" (Barofsky, 1986).

Soon thereafter, sociologists and economists led the efforts to define and

measure the term "quality of life." Today, theoretical and empirical

arguments continue to be waged about the definition, measurement, and

implementation of quality of life concerns by a host of academicians and

professionals including sociologists, economists, psychologists, and

physicians, as well as insurance companies, environmental groups, and

public policy contributors. In the health care arena, the emergence of life-

saving medical technology has increased the motivation to measure quality

of life because these breakthrough treatments may increase an individuals'

length of life while not necessarily maintaining or enhancing the quality.

This impetus to successfully measure quality of life, however, has not led to

any quick answers.

Despite multi-disciplinary efforts to define and measure quality of life,

it is apparent that there is no agreed upon definition or operationalization for

quality of life. The term "quality of life" has been said to mean all things

that one might want to measure beyond the traditional outcomes of death or

disease activity (Guyatt, Feeny, & Patrick, 1991). Other researchers have

suggested that quality of life measures the difference, or gap, at a particular

period of time between the individual's hopes and expectations and that

individual's present experiences (Calman, 1984). Although no agreement

about the definition of quality of life seems to exist, most health care

professionals agree that quality of life is a critical variable to consider in life-

saving treatments.

Measurement Approaches

The major measurement distinction in quality of life research has been

whether to measure it as a single or multi-dimensional variable. Proponents

of the single dimensional approach simply ask patients: "How is your quality

of life?" They argue that rating one's own quality of life is a daily pursuit

when someone asks, "How was your day?" In a sense, the question taps

the quality of a person's life during that particular day. In contrast, the

multi-dimensional proponents have suggested that quality of life

encompasses a variety of domains that can form a composite quality of life

index. These include such areas as activity level, daily self-care, overall

health, marital and family relations, career, spiritual, and religious

satisfaction. Proponents of the multi-dimensional approach argue that these

domains must be measured individually in order to properly sample the

components of one's perception of his/her quality of life. Schipper and

Levitt (1986) have characterized the current conflicts between global and

multi-dimensional efforts by stating, "the more global the definition, the

more abstract, and the more abstract, the less measurable." The authors

concluded "the less measurable, the more scientifically uncomfortable"

(p.20) scientists and researchers become. Presumably because of

arguments such as this, many researchers have tended to be more

comfortable with physiological or biological outcome measures (e.g. left

ventricular ejection fraction) because they are seen as more scientifically

rigorous and reproducible. Nonetheless, researchers have implicitly

continued to measure similar core attributes thought to be associated with

quality of life, such as disease symptoms and/or treatment side effects,

functional status, psychological distress, social interaction, sexuality and

body image, and satisfaction with medical treatment (Aaronson, 1986).

MacKeigan and Pathak (1992) recently summarized the practice of

measuring the multiple domains of quality of life by tapping into the "5 D's":

death, disease, disability, discomfort, and dissatisfaction. They further

group these health outcomes into mortality (death) and morbidity (disease,

disability, discomfort, and dissatisfaction).

Guyatt, Feeny, and Patrick (1991) have suggested that a resolution to

the differing views of quality of life measurement is the inclusion of multiple

indices that represent both the uni-dimensional and multi-dimensional

paradigms. They suggest that the advantage of multiple measurement is

that specific information about a particular domain of interest such as

mobility or social activity can be obtained. The more general measurement,

in turn, would provide information that may be more generalizable and of

interest to policy makers or third party payers. The disadvantages of

multiple means of measurement would include potential burden to subjects

and possible interpretive difficulties if improvement is shown by some

measures but not on others. Table 1 provides a brief listing of the various

types of quality of life measures currently available for health outcome


Perhaps the most practical approach to measurement of health-related

quality of life has evolved from the early studies of the Health Insurance

Experiment funded by the RAND Corporation that sought to measure

changes in health, from children to the elderly and from health to illness

(Brook, Ware, & Davies-Avery, 1979). Building on the knowledge gained

from these early attempts of measuring health, a group of researchers

sought to examine the multiple indices of health in a large sample (N =

22,462) at over 500 sites in three large cities with the purpose of deriving

scales that would meet the diverse definitions and needs of researchers

seeking to measure and track changes in health status. The products of

these efforts have been a set of scales that measure distinct components of

health using both rational and empirical scale development techniques.

Thus, the Medical Outcomes Study Health Survey Short Form-36 (SF-36)

was derived. The SF-36 assesses 8 domains of health functioning including:

physical functioning, role limitations due to physical problems, role

limitations due to emotional problems, social functioning, bodily pain, mental

health, vitality, and general health perceptions. Descriptions of each of the

scales is available in the Method section. The items and scales of the SF-36

are scored so that a higher score indicates a better health state. Initial

psychometric information appears to be encouraging, with appropriate

reliability and validity having been established.

Table 1

Quality of Life Measures

Type of Measure

Definition of Measure


Mental Status Psychological and Mini Mental Status Exam
psychiatric testing
intended to assess
adaptation, personality,
and/or mood.
Functional Status Measures person's ability New York Heart
to perform activities of Association rating
daily living (ADL's) or
other physical abilities.
General Health Status Intended to measure McMaster Health
overall personal health Questionnaire
and the absence of
Symptom Checklists Measures the frequency Transplant Symptom
and severity of somatic Frequency and Distress
symptoms. Some Scale
measures are available in
disease-specific formats.

Life Satisfaction Intended to measure Satisfaction With Life
individual's perceived Scale (SWLS)
contentment with life.
Psychosocial Adjustment Measures the multiple Psychosocial Adjustment
components that form a to Illness Scale (PAIS)
composite score of
overall quality of life.
Global Quality of Life Intended to quantify EuroQoL
quality of life into a
single quotient.
Multiple-Domain Health Measures the multiple Short Form-36 Medical
Status components of health Outcomes Study Health
including physical Survey (SF-36)
functioning, mental
health, vitality, and role

The differing views about the conceptualization of quality of life have also

resulted in different assessment modalities such as self-report questionnaires

and clinical interviews. Mayou (1990) suggested that although

questionnaires provide quick assessments of quality of life, they may not be

as sensitive to individual differences as other measurement techniques. For

example, a semi-structured interview would permit more open-ended

responding that can allow for both quantitative and qualitative analysis. A

brief structured interview would also allow various health professionals to

assess an individual's quality of life accurately and efficiently. It is

important to note that the reliability of quality of life assessments by

questionnaire and interview has been examined (Bremer & McCauley, 1986)

and results indicated that reliability of measurement was not affected by

location or format. More clearly, the researchers varied the quality of life

assessments by whether subjects completed the measures at home or in the

hospital and by interview or questionnaire. Their results showed equivalent

reliabilities regardless of the setting or mode of response, suggesting that

either interview or questionnaire provide consistent quality of life

measurement. The validity of the interview could be further established by

considering not only the patient's view but also the spouse/significant

other's perception of the patient's quality of life.

Mayou (1990) recently critiqued both the conceptual and

measurement paradigms of quality of life in the cardiovascular disease

literature. He listed the overemphasis on general quality of life as a

particular problem, suggesting that general measures neglect important

aspects of quality of life such as mental status, individual satisfaction, and

family/significant other satisfaction. Mayou (1990) stated that quality of life

should be defined in terms of what is important to the family and "must

include a measure of satisfaction" (p. 101). He pointed toward

methodological problems such as the reliance on self-report measures rather

than interview-based measures and the use of unrepresentative samples.

Following the recommendations of various researchers (Guyatt, Feeny,

& Patrick, 1991; MacKeigan & Pathak, 1992), the current study has chosen

to employ a multiple measurement paradigm designed to sample from the

best of the existing instruments. Quality of life, for the purposes of this

study, was simply a general label referring to a lifestyle in which the patient

is relatively disease symptom-free, highly satisfied with life, and able to

vigorously function with few limitations due to physical or mental


Predictors of Quality of Life

Partial support for hypothetical relationships can be gleaned from

research in other areas of transplantation and medical procedures that were

designed with similar goals of prediction of post-intervention outcome.

Although generalizations from such studies should be made cautiously, a

brief review of this literature is presented.

Depressive disorders (major depression, organic affective disorder,

dysthymic disorder, and adjustment disorder with depressed mood) are

generally agreed upon to be a common co-occurrence of significant medical

conditions (Beidel, 1987). The assessment and intervention in depressive

symptomatology in transplantation has received the most attention from

health psychological researchers. Maricle et al. (1989) reported an incidence

rate for depressive syndromes in cardiac transplantation candidates of 54%

(37 of 68). Furthermore, Maricle et al., reported that depressive

symptomatology was associated with demographic factors such as younger

age, less education, lower occupational achievement and previous

psychiatric history.

Although depression is widely recognized as a potentially concomitant

difficulty for transplantation candidates and recipients, the exact impact on

post-surgical outcome remains somewhat unknown. The relationship

between depression symptomatology and survival has been investigated.

Maricle, Hosenpud, Norman, Pantley, Cobanoglu, & Starr (1991) examined

the predictive validity of preoperative psychological distress, as measured by

the SCL-90-R, on post-transplantation graft rejection and infection rate in a

sample of 58 cardiac recipients. Their results indicated that there was no

relationship between the global symptom index and graft rejection and

infection rate. However, follow-up analyses without Bonferronni correction

indicated that the depression subscale of the SCL-90-R was significantly

associated with infection (r = -.27). Significant associations were also

found with rejection in which anxiety (I = .31), psychoticism (L = .30), and

positive symptom distress ( = .29). These correlations were not significant

after correcting for the multiple hypotheses. The authors concluded that the

SCL-90-R had only limited predictive validity in post-transplantation medical

variables. However, the authors failed to acknowledge that psychological

factors would be expected to be more clearly implicated in quality of life

outcome measures.

Colon et al. (1991) also studied depression in 100 acute leukemia

patients who were undergoing allogenic bone marrow transplantation.

Results indicated that patients who reported depression as a prominent

symptom at the pre-transplantation evaluation had poorer 1-year survival

rates following transplantation. Additional results indicated that patients

who were rated as having a high level of perceived spouse/family support

were found to have improved survival when compared to patients with low

or moderate support. The researchers concluded that depressed mood was

a negative "marker" for outcome in bone marrow transplantation patients.

In contrast, Maricle et al. (1989) found no relationship between depression

and mortality in a small sample (N = 43) of cardiac transplantation


Depression has also been examined as a negative predictor of quality

of life. Rodin and Voshart (1987) studied a group of renal transplantation

candidates (N = 115) classified as depressed on the BDI (scores > 17).

Results indicated that depression scores were significantly associated with

lower general quality of life (r =.59) and more unemployment, even after

controlling for relevant demographic factors (Rodin & Vorhart, 1987).

Maricle et al. (1989) found a significant trend in long-term cardiac

transplantation survivors between pre-operative depression symptomatology

and employment such that 60% of the non-depressed group had returned to

work while only 28% of the depressed group were employed. Although

moderators of the depression-employment relationship probably do exist, no

examination of potential moderators was proposed in this study.

Other studies have begun to investigate potential moderators of the

relationship between depression, survival, and quality of life. Conn, Taylor,

and Wilman (1991) hypothesized that depression may impact quality of life

by depleting energy needed for active coping efforts. The authors studied

the impact of mood states on quality of life in a group of coronary artery

bypass graft (CABG) patients. Their results indicated that depression scores

accounted for 49% of the variance in quality of life scores, however, no

relationship was found between anxiety and quality of life. Depression

scores were also strongly related to decreased exercise, a poor diet,

increased medications, stress, and smoking.

Another interesting hypothesis regarding the presentation of

depressive symptoms involves the idea that patients that are the "sickest"

from a cardiological perspective would show the most psychological

distress. Preliminary studies of this hypothesis in cardiac transplantation

candidates, however, do not support this idea. Maricle et al. (1991) found

no replicable relationship between cardiac function and report of depressive

symptoms. In fact, in a second study, their results indicated that patients

that reported the most depressive symptomatology had significantly better

cardiac indices. These findings suggest that distress is associated with

patients who have improved cardiological output. These findings are

consistent with another study of cardiac patients in which higher anxiety

was associated with better cardiac output (Schocken, Greene, Worden,

Harrison, & Spielberger, 1987).

Psychosocial variables, such as family and social support, are also

widely believed to have an important impact on the ultimate medical and

psychological success of the transplant. Christopherson and Lunde (1971)

reported that in a sample of cardiac transplantation recipients, open patterns

of family communication that supported the decision to undergo

transplantation were associated with positive post-transplantation

adjustment. Mishel et al. (1987) has suggested that transplantation families

go through a process of "redesigning the dream" which requires the family

to continually adjust to the changing demands and goals of post-

transplantation life. The family's ability to modify and ultimately renew the

roles of the transplantation recipient may be related to the recipient's

perceived quality of life. Baker, Curbow, and Wingard (1991) studied the

relationship between post-bone marrow transplantation role retention and

quality of life. Role retention was defined as the degree to which the

recipient could perform the family, work, and community roles that they

were accustomed to performing. Results indicated that role retention was

significantly associated with measures of quality of life including current and

future life satisfaction and measures of positive affect. Furthermore,

significant relationships were found between negative mood and role

retention. These findings suggest that negative mood affects quality of life

by reducing the roles of the transplantation recipient.

Other dimensions of negative affect, such as anxiety and anger, have

received significantly less attention from transplantation researchers than

depression. This is somewhat surprising given the moderately high

correlations between the constructs of depression, anxiety, and anger which

have led to suggestions that a single factor of negative affect may underlie

these overlapping dimensions (Matthews, 1988). Nonetheless, anxiety has

been implicated as an important presentation of distress during critical

stages of the transplant process (Kuchler et al., 1991) and may serve as an

indicator for psychotherapeutic intervention during these periods. Anxiety

has been suggested as a possible hindrance in self-care activities that may

indirectly influence quality of life (Conn, Taylor, & Wilman, 1991).

However, this hypothesis was not supported in a sample of CABG surgery

patients. Similar hypotheses about the self-report of anger have been found

in hypertensive patients in which self-report of anger was found to be

associated with medication non-compliance (Lee et al., 1992). Nonetheless,

anxiety and anger may also serve as a potential obstacles in information-

gathering about the transplantation process that requires the patient to

engage in emotion-focused coping rather than problem-focused coping

(Lazarus & Folkman, 1984). These emotion-focused coping strategies could

subsequently leave the recipient unprepared to independently deal with the

rigors of compliance with the medical regimen in the post-transplantation


Although negative affect could be seen as potentially problematic for

the transplantation patient, alternative views also exist. For example,

Rabinowitz and Van Der Spuy (1978) found that the admission of anxiety

and hostility were associated with positive adjustment in renal

transplantation patients. This data is consistent with the view that patients

that exhibit anxiety and hostility may be more actively confronting the exact

seriousness of their medical condition and therefore, more efficiently dealing

with their feelings and subsequently adapting to their situation. As Beidel

(1987) has suggested, anxiety and angry feelings remain to be examined

more systematically across the stages of transplantation.

Quality of Life by Proxy

Perceived quality of life may vary by the perspective of the rater.

Different raters are interested in sampling different domains of quality of life.

For example, the medical professional may be interested in the surgical

outcome and restorative function of the replaced organ, while the

psychologist may be interested in the person's psychosocial functioning

post-transplantation. Patients may be particularly interested in the cosmetic

aspects of procedures and changes in the inter-familial relationships or

sexual functioning due to a procedure. Still another perspective is available

through the eyes of the patient's spouse or significant other who may be

interested in such things as degree of independence. Mishel and Murdaugh

(1987) have suggested that the impact of transplantation on the family

requires significant role change within the relationship between recipient and

the spouse. The role of the spouse and the caregiver has gained increased

attention from researchers. Sechrest and Pitz (1987) commented that it

was "impossible" to examine fully the issue of quality of life without a

simultaneous discussion of a transplant's impact on the recipient's

significant others.

The similarity between spouse and patient views has been examined.

Gray, Brogan, and Kutner (1985) sampled 50 end-stage renal disease

patients' satisfaction and abilities in 13 life areas such as leisure time,

working life, sexual functioning, and social life and compared their responses

with their spouse's own self-view. Interestingly, the patient and spouse

views were congruent on 10 dimensions. Patients and spouses differed on

satisfaction with medical care, patient's perception of self care, and

satisfaction with feelings about self in that patients reported lower levels on

these dimensions. This data suggests that there is a high correspondence

between patients and their spouses in perceptions of important life areas.

The concerns of spouses of transplantation recipients have also been found

to be consistent with one another. For example, Buse and Pieper (1990)

studied the perceptions of 30 spouses (26 women & 4 men) after their

spouse received a cardiac transplant during the pre- and posttransplantation

periods and found that cardiac spouses do not experience a reduction of

stress from pre to post-transplantation but rather maintain high levels of

stress. More specifically, the central concerns of spouses appeared to be

the possible loss of the loved one, life in general, and the need to learn more

about the transplant.

The spouse's perspective may be particularly useful, because it is very

likely that it is the spouse who spends the most time with the patient, due

to the extensive care-giving responsibilities. However, care-giving has been

shown to have a significant impact on the caregiver. Caregiver research has

recently been flourishing, particularly in the care of the elderly and has

focused on two constructs: caregiver burden and caregiver strain.

Caregiver burden refers to the more tangible and concrete aspects of

providing services, while caregiver strain refers to the subjective appraisal of

stress that providing care has on the caregiver (Parks & Pilisuk, 1991).

Caregiver strain has received the most research attention and such factors

as work interference, personal life impact, and care-recipient impairment

have been shown to be significant predictors of reported strain (Scharlach,

Sobel, & Roberts, 1991). Women also appear report more caregiver strain,

probably because they are more likely to involve themselves in the role of

the caregiver (Parks & Pilisuk, 1991). The degree of caregiver strain is

thought to be part of the view that a proxy rater would have of the patient.

The issue of rating a patient's quality of life by proxy has begun to

draw the attention of researchers because of its potentially rich source of

easily attainable data. Magaziner et al. (1988) studied the comparability of

361 elderly hip fracture patients and their spouses interview responses to

measures of health and functional status. The results indicated that proxies

tended to overestimate patient disability relative to the patients themselves,

especially with regard to the capacity to complete activities of daily living.

Proxies with the greatest contact with the patient tended to have the

highest agreement with patient responses. However, on measures of

functional and health status, proxies tended to report more patient disability

than the patient. The authors provided tentative reasons for proxy

overestimation such that proxies may want to elicit sympathy and additional

help from others or it may be a function of their own over-protectiveness of

the patient.

Epstein et al. (1989) found similar results in an adult sample of

primary care patients at a university medical center. The researchers asked

proxies to respond to measures of functional, emotional and health status,

social activity, and satisfaction in the same manner that they thought the

patient would. Patient and proxy correlations were highly significant for

overall health, functional status, social activity, and emotional health and

moderately correlated for satisfaction. Proxies reported lower satisfaction

and emotional health than did the patients. Proxies who spent more time

per week helping the patient rated the patient as more impaired than did the

patient. The authors concluded that proxy responses are not able to inter-

mixed with patient response, rather proxy responses represent a different

data set that should be examine separately.

Sprangers and Aaronson (1992) provided an overview of the research

studies comparing patient and significant other ratings of health status and

quality of life. Their review of 13 studies suggested that the level of

agreement between patients' self-reports and significant other's ratings was

dependent on the concreteness, visibility, and subjectivity of the quality of

life dimensions assessed. The authors found that the less concrete and

visible and the more subjective the information, the lower the concordance

rates. Ratings of such things as functional status seem to be highly

correlated while ratings of psychological distress seem to be correlated

moderately or minimally. The authors concluded that there is a tendency for

significant others to more negatively rate the quality of life and performance

of the patient. Some evidence exists that variables such as proximity of the

raters living arrangements (e.g. in the same household), performing a

caregiving function, and amount of time devoted to caregiving impact the

level of correspondence between significant other and patient ratings.

The central aim of many of the studies in the quality of life by proxy

area was to examine the correspondence between patient and spouses in

order to replace missing values of patients who either refused or were

unable to respond to the measures of the given study. Although this is a

worthy goal for research methodology, the present study intended to sample

the significant other's perspective for its own sake. This strategy was

designed to capitalize on the unique perspective that the significant other

holds on the patient's quality of life. The two sets of ratings were assessed

for correspondence but no replacement of missing data was performed.

Instead, parallel analyses were performed in order to find significant markers

for both the patient's and the significant other's ratings of quality of life.

Quality of Life in Cardiac Transplantation

Quality of life measures have been used in the assessment and follow-

up care of cardiac transplantation recipients in a variety of studies during the

last decade. The following section provides a detailed analysis of the

current literature examining quality of life in cardiac transplantation

recipients and concludes with a brief summary of critiques and findings from

these studies.

Review of studies

Quality of life has been operationalized in a variety of ways. Some

studies have examined quality of life by using simple questions about the

perceived quality of life in multiple domains of interest. Bunzel, Grundbock,

Laczkovics, Holzinger, and Teufelsbauer (1991) surveyed 35 patients (32

male, 3 female) about their postoperative progress and satisfaction on visual

analog scales of quality of life. The researchers defined life quality in nine

areas: physical, emotional, mental, vocational, and sexual status, financial

situation, leisure activities, partnership, and overall life quality. The patients

sampled ranged from 13 to 45 months post-transplantation. Results

indicated that patients reported increased functioning and satisfaction in all

9 areas with the exception of financial status. However, the authors

acknowledged that this methodology allows only for an approximation

because of the limitations of retrospective research. More clearly, no pre-

transplantation data was available for systematic comparisons. The authors

asserted that patient satisfaction with a medical procedure may be the most

important result of a medical procedure and may be highly associated with

increased life quality.

Simple quality of life questionnaires have also been combined with

health indices in an attempt to predict quality of life by the relative

contribution of somatic complaints or symptom distress. Lough, Lindsey,

Shinn, and Stotts (1987) developed a symptom checklist designed especially

for common symptoms reported by cardiac transplantation recipients. In

their study, a list of 29 symptoms was mailed to 100 cardiac recipients with

instructions to rate the frequency and emotional distress caused by each

symptom on a five-point scale (0 = never have that symptom to 4 =

always have that symptom; 0 = not at all upsetting to 4 = extremely

upsetting). The sample of returns included 65 men and 10 women. The

time since transplantation ranged from 7 months to 14 years. Although

medication types were controlled for and some significant differences

occurred between medication groups, there were some symptoms that

occurred "often" or "always" regardless of medication type. The symptoms

that were most common included: changed facial appearance, changed

bodily appearance, pain, overeating, fatigue, lack of sleep, and decreased

interest in sex. The most common symptoms were not always the most

distressing; sexual difficulties and decreased sexual interest were rated as

the most distressing. Changes in facial and bodily appearance and mood

swings also ranked highly among the most distressing symptoms. The

cardiac recipients generally described their quality of life as good to excellent

(89%), with 82% reporting good to high levels of satisfaction with their

current quality of life. Perhaps the most significant finding from this study

involved the regression analysis that indicated that perceived quality of life

was significantly predicted by satisfaction, accounting for 63% of the

variance. Symptom frequency and symptom distress accounted for a

combined 4% of the variance in quality of life. The authors concluded that

while transplantation recipients experience recurring physical symptoms and

associated emotional distress, other life factors have greater impact on

perceived quality of life. These findings indicates that symptom reports are

important because symptoms can be distressing to the recipient. However,

the report of symptoms appears to distinct from traditional quality of life


Other researchers have used test batteries designed to tap the various

psychosocial dimensions presumed to be important. Jones, Chang, and

Esmore et al. (1988) contacted 38 consecutive cardiac transplantation

recipients for follow-up psychological and quality of life assessments.

Subjects completed the Beck Depression Inventory, the State-Trait Anxiety

Inventory, the Campbell Well-Being Scale, and measures of body image,

current marital and occupation functioning, and perception of

transplantation. Patients were evaluated before transplantation, at discharge,

and at the 4-, 8-, and 12-month post-transplantation intervals, but the exact

number of patients at each of the follow-up assessments varied. All 38

patients were assessed at their first hospital discharge while 32 were

assessed at 4-months, 21 at 8-months, and 14 at 12-months. The initial

sample included 32 men and 6 women who were generally middle aged (X

= 38.8 years old), married (79%) with moderate levels of education (X =

10.8 years). Results indicated that patients reported significantly higher

levels of depression and state anxiety prior to transplantation compared to

post-transplantation. Consistent with the psychological measures, the

measures of well-being and body image indicated significant improvements

from pre to post-transplantation. In addition, it was found that 64% of the

patients had either returned to full-time work or to home duties by the time

of the 12-month follow-up.

More recently, Jones, Taylor, Downs, and Spratt (1992) followed-up

on 38 consecutive cardiac transplantation recipients from the Jones et al.

(1988) study to assess current quality of life. Of the 38 in the initial

sample, 27 patients were available and participated (6 patients died, and 5

were unable to be contacted) in a long-term follow-up approximately 4 years

post-transplantation. Each patient completed the pre-transplantation battery

as well as the Nottingham Health Profile (NHP), which is a measure of pain,

physical mobility, sleep, energy, social isolation and emotional reaction.

Consistent with the previous study, significant differences were found on

measures of depression, anxiety, and well-being between pre and post-

transplantation. However, no significant differences were found between

any of the follow-up periods. More clearly, the initial gains post-

transplantation did not increase significantly at the subsequent time periods.

However, it should be noted that the well-being scores at each of the follow-

up periods were significantly higher than the general population, suggesting

that transplantation recipients may view their lives with more positive

meaning following transplantation than the general population. On the NHP,

comparisons of the recipient's scores to a general medical population

suggested that the recipients reported less pain, more energy, and less

emotional reactions. Overall, these studies by Jones and his colleagues

show consistent quality of life and psychological benefits for the cardiac

recipient immediately following transplantation longitudinally up to a 4-year


Other batteries to examine the psychosocial functioning have been

developed by Freeman, Folks, Sokol, and Fahs (1988). These authors

evaluated a group of 70 cardiac transplant recipients during the pre-

transplantation stage as well as in regular follow-ups post-transplantation

(within the first 6 months, again between 6 and 12 months, and annually

thereafter). Subjects were administered a Mini-Mental Status Exam (MMSE),

the State-Trait Anxiety Inventory, the Zung Depression Scale, and the

Psychosocial Adjustment to Illness Scale (PAIS), a measure of the patient's

adjustment in 7 domains: health concerns, vocational status, domestic

environment, sexual functioning, family, social activity, and psychological

distress. The sample included 50 men and 20 women and was 84%

Caucasian and 11% African American. The authors presented some general

baserates of psychiatric diagnoses and interventions. Pre-transplantation

assessments indicated that 26 patients were given Axis I psychiatric

diagnoses while 8 were given Axis II diagnoses. Results indicated a

significant relationship between an Axis I diagnosis and persistent regrets

following surgery. The authors reported that 24 out of the 70 recipients

assessed required a psychological intervention of some kind. The

longitudinal comparisons from pre to post-transplantation indicated

statistically significant improvements in anxiety, depression, and adjustment.

However, no differences were found on the MMSE. Although disease type

was not a focus of this study, the researchers noted that more difficulty

with adjustment was observed in individuals with acute onset

cardiomyopathy, such as that seen in viral cardiomyopathy in which heart

failure can occur in less than one year. It is thought that the acute nature of

disease onset allows the patient less time to adjust to his/her condition,

given that prior to this time the individual may have been relatively disease-


Some researchers have utilized the Nottingham Health Profile as an

index of quality of life. O'Brien, Banner, Gibson, and Yacoub (1988)

implemented the NHP in pre and post-transplantation assessments with 48

heart and lung patients. Their sample included a majority of women (67%)

with a mean age of 23.8 years. The NHP was administered by interview at

assessment and by questionnaire at the 3, 6, and 12-month periods post-

transplantation. The results indicated significant improvements in each of

the areas of Part I of the NHP including energy, pain emotional reactions,

sleep, social isolation, and physical mobility. Similar significant

improvements were found in Part II of the NHP including the domains of

occupational, domestic, social, sexual, and recreational (hobbies and

holidays). The authors noted that the NHP was particularly useful in

combining both clinical and research applications because it was easily

adapted as either a questionnaire or as an interview instrument.

Caine, Sharpies, English, and Wallwork (1990) used the NHP to

compare within subjects changes over the stages of cardiac transplantation.

Patients (n = 196) responded to the NHP at regular intervals following

acceptance into the transplantation program. All patients completed the

NHP immediately after acceptance to the waiting list prior to transplantation,

and regularly at 3 month intervals prior to transplantation and throughout the

post-transplantation period up to 5 years. No data was provided indicating

the number of subjects reporting at each interval. Short-term follow-up

results from comparisons of the most recent pre-transplantation reports to

the 3-months post-transplantation evaluation indicated significant

improvements in all six dimensions for the 122 patients sampled. However,

similar to other studies (Jones, Taylor, Downs, & Spratt, 1992), no

statistically significant improvements were found in comparisons between

the 3-month post transplantation reports and any following post-

transplantation reporting period. All post-transplantation reports were within

normal limits when compared to a normative sample of adults. The

researchers also found that the NHP was a reliable predictor of outcome in

terms of length of stay in the hospital and return to work and leisure activity

at 3 months post-transplantation.

Because most quality of life research indicates that quality of life

does, indeed, increase as a result of cardiac transplantation, a logical step

for researchers would be the attempts to find significant pre-transplantation

indices that are predictors of post transplantation quality of life (Shapiro,

1990). Brennan, Davis, Bucholz, Kuhn, and Gray (1987) conducted a pilot

study with 11 transplantation recipients that examined pre-transplantation

DSM-III diagnoses, the intervening medical course, and quality of life at 9 to

15 months post-transplantation. The authors performed psychiatric

evaluations on all patients and diagnosed 3 patients with personality

disorders, 3 with depression/anxiety, and 3 with dementia/delirium

symptomatology. Medical course was operationalized by the number of

days spent in the hospital following the transplant and the number of

biopsies performed. It was presumed that the number of biopsies performed

during a given period was related to the severity of rejection undergone

during that period. The average number of biopsies performed was 10.5

during the first six months, and 3.4 during the second six months. Results

showed that patients with more positive quality of life following cardiac

transplantation tended to be married, did not carry a diagnosis of a

personality disorder before the operation, had a shorter post-transplantation

hospital stay, had fewer days of readmission, and underwent fewer biopsies

in the year following cardiac transplantation. The authors suggested that

there may be some inter-relationship between marital status and personality

disorder influencing medical outcome, which in turn influenced quality of life.

The authors concluded that compliance to the medical regimen may also be

a moderating factor.

Mai, McKenzie, and Kostuk (1990) also studied the role of

psychopathology and compliance in a study of the pre to post-

transplantation psychosocial and quality of life adjustment of 27 consecutive

cardiac transplantation recipients. Subjects responded to the Present State

Examination (PSE), a structured psychiatric interview schedule, the

Symptoms Checklist (SCL-90), and the General Health Questionnaire (GHQ).

In addition, the clinical team rated the patients on a 3-point scale of

compliance (good, fair, poor) that was based on the self-report of the patient

as to the extent to which he/she had followed medical requirements

concerning diet, salt restriction, smoking, exercise and drug and alcohol

ingestion. The medical team also completed a non-standardized quality of

life measure that assessed current physical activity, employment status, and

sexual activity from the medical record. At the pre-transplantation

assessment, 14 of the 27 patients received a PSE psychiatric diagnosis,

while 5 of the 24 received a PSE diagnosis at the 12 month post-

transplantation follow-up. An association was found between the pre-

transplantation rating of compliance and psychiatric diagnosis. Pre-

operatively, 47% of patients rated as compliant had a psychiatric diagnosis,

whereas 100% of the patients rated as having "fair" or "poor" compliance

had a psychiatric diagnosis. The authors hypothesized that the presence of

psychopathology impairs the capacity of the individual to follow the medical


Additional psychosocial information was gained from the

psychometric assessments that utilized questionnaires. Results from the

SCL-90 indicated highly significant improvements from pre to post-

transplantation scores on the Anxiety, Phobic Anxiety, Additional Items,

Total Score, and Symptom Distress Index. In contrast, the GHQ did not

show statistically significant pre to post-transplantation improvement.

However, trends were noted in the expected direction of improvement.

Statistically significant improvements were also shown in the areas of

physical activity and employment indicating that most patients had returned

to a desired level of activity. The sexual activity measure was only able to

be compiled for 14 subjects and the data suggested that libido was normal

in 8 patients, impaired in 3, and absent in 3. Overall, the authors indicated

that patients did increase their quality of life after transplantation because of

the significant improvements noted in physical activity, return to work, and

resumption of sexual activity.

Packa (1989) contacted 22 cardiac transplantation recipients

approximately 6-12 months after transplantation. Each of the subjects

completed the McMaster Health Index Questionnaire (MHIQ) and the Cantril

Self-Anchoring Scale. Results from the MHIQ indicated that subjects

reported "satisfactory" levels of functioning in the physical, social, and

emotional domains. The Cantril Self-Anchoring Scale was administered by

asking subjects to rate their quality of life on a continuum (0 = worst

possible quality of life imaginable and 10 = best possible quality of life

imaginable). Subjects were asked to make ratings of their quality of life

retrospectively at one-month before transplantation and at time of interview.

In addition, subjects were asked to forecast their anticipated quality of life at

5-years post-transplantation. Significant differences were found between

the scores at all three time periods such that cardiac transplantation

recipients expected a continued progression of improvement over time.

Summary and Conclusions

Before drawing conclusions about findings from the studies reviewed,

several caveats are necessary. First, each of the studies generally relies on

convenience sampling at the researcher's particular setting. The samples

tend to be consecutive patients that are evaluated pre-transplantation and

are subsequently deemed as "acceptable" for the transplantation waiting list.

Thus, patients are not randomly selected, but rather are purposefully

selected based on both medical and psychosocial characteristics. Second,

the samples tend to be relatively small in number. Although the number of

cardiac transplantation centers is rapidly growing, each center generally does

not transplant large numbers of patients and- multi-center studies are the

exception rather than the rule (Paris, Woodbury, Thompson et al., 1992).

Third, great variability in the operationalization of the constructs of interest

is evident. The quality of life construct certainly has a wide range of

meanings and relatively little consensus can be found about how to measure

it best. Fourth, researchers have traditionally relied only on the self-report of

the patient of his/her quality of life and have not compared the alternative

views of significant others such as the spouse or extended members of the

family. Fifth, the timing of assessment and the number of assessments also

varies dramatically across studies providing a mosaic of views of quality of

life at different stages and times of post-transplantation. Despite these

shortcomings, the critical work related to quality of life post-transplantation

continues while each researcher attempts to define or operationalize quality

of life according to his/her purposes.

Although it is premature to form conclusive statements about the

cardiac transplantation quality of life literature, general trends can be noted.

The studies reviewed above suggest that the quality of life of cardiac

transplantation recipients increases after undergoing transplantation. The

changes are statistically significant from pre- to post-transplantation and are

in place within 3-4 months following surgery. Longitudinal repeated

comparisons of quality of life after the 3-4 month period are generally not

statistically significant, suggesting that most of the significant quality of life

change occurs in the period directly following transplantation (Jones, Taylor,

Downs, & Spratt, 1992). Additional studies have indicated that reports of

anxiety and depression are significantly reduced from pre- to post-

transplantation (Jones et al., 1988). Concerns about sexual functioning and

appearance appear to be frequently reported, post-transplantation (Lough et

al., 1987). The return to work rates appear to be promising ranging from

56% to 64% (Shapiro & Kornfeld, 1989), but no systematic predictors have

been established (Jones et al., 1988; Paris et al., 1992). Because most

quality of life research indicates that quality of life does, indeed, increase as

a result of cardiac transplantation, Shapiro (1990) has suggested that

research should focus on establishing significant pre-transplantation

predictors of post transplantation quality of life.

Previous Research at the University of Florida

The psychological perspective on cardiac transplantation has a

relatively strong history at the University of Florida. Cardiac transplantation

candidates have routinely been evaluated by the faculty and students of the

Department of Clinical and Health Psychology since the commencement of

transplantation at Shands Hospital in 1985. Psychologists have been

consulted as part of multi-disciplinary teams presumably to make

assessments of the candidate's psychosocial situation by sampling the

patient's mental status, level of affective distress, substance use/abuse

history, medical compliance history, coping style, social support, and

knowledge of current illness. In the context of these evaluations, clinical

interviews and psychological testing were implemented to aid in the clinical

management of patients and a complementary research protocol was

established. The evaluations were designed to broadly sample the patient's

affect, behavior, and cognition. Accordingly, psychological tests were

selected that would best measure the domains of interest. The exact

measures for each of these domains have undergone some changes along

with the personnel performing the evaluations. Nonetheless, some measures

have remained constant and those measures are the focus of the current

study. Personality has been assessed using the Minnesota Multiphasic

Personality Inventory (MMPI) (Hathaway & McKinley, 1967), which has been

called the "most thoroughly researched personality assessment instrument

available" (Green, 1985, p. 318). The MMPI is a self-report psychiatric

screening measure that provides scores on numerous subscales and allows

for examination of response-style distortions through the use of "validity

scales." Intellectual and cognitive status have been examined through the

use of the Wechsler Memory Scale (WMS) (Wechsler, 1945) and the two

subscales, Vocabulary and Block Design from the Wechsler Adult

Intelligence Scale-Revised (WAIS-R).

The clinical practice of routine assessment of cardiac transplantation

candidates has ended itself well for data collection for research. For

example, Hecker, Norvell, and Hills (1989), while working at the University

of Florida, were the first authors in the psychological literature to validate

the use of MMPI profiles in the assessment of cardiac transplantation

candidates. The authors also established a normative data base for cardiac

transplantation candidates on other psychological measures. They

performed cluster analyses and found three MMPI profile subtypes that have

been replicated (see below, Robinson, Greene, & Geisser, 1993). Cluster 1

("Neurotic Triad" and/or "High Distress") included patients who were

reporting a variety of physical complaints, were significantly anxious and

depressed, and admitted to a degree of confusion. Cluster 3 ("Subclinical V-

type") also reported significant physical complaints but did not report

significant depression or anxiety. Finally, Cluster 2 ("Normal") reported no

significant elevations on MMPI scales and were indicative of a fairly well-

adjusted group. Caution in interpretation was emphasized by the authors

because the analyses included only 64 patients.

More recently, Robinson, Greene, and Geisser (1993) also used

cluster analyses to examine MMPI profiles in various medical populations

including chronic lower back pain, headache pain, and cardiac

transplantation candidates. The cardiac disease population included both the

Hecker et al. (1989) sample and the sample that will be used again in the

current study. These authors found four clusters in a large sample of

cardiac transplantation candidates including the 3 clusters that were

originally found by Hecker et al. (1989). The fourth cluster group appeared

to form with members of original Cluster 1 that separated into two groups:

the "Neurotic Triad" and the "High Distress" groups. It also important to

note that the "Normal" group remained composed of subclinical scores on all

MMPI scales but showed a near elevation on Scale 9 (Hypomania scale).

Although both Hecker et al. (1989) and Robinson et al. (1993) found similar

profile groups, the predictive validity and utility of the clusters in the

assessment of post-operative functioning remain to be examined.

Longitudinal assessment of the cardiac transplantation candidates has

been examined in two investigations. First, a case study has been presented

that demonstrated the clinical utility of psychometric testing across the

stages of transplantation (Greene, Sears, & Mills, 1993). Second, a

repeated measures design was employed to assess cognitive changes

immediately post-transplantation (2 weeks) and after one-year. Personality

was also assessed at the one year point. Results indicated at post-

transplantation that measures of depression were significantly reduced at the

post-transplantation period. In addition, a trend was noted for memory

functioning such that recipients tended to increase their ability to perform

memory tasks as operationalized by the memory quotient. The repeated

measures findings, however, were limited by a somewhat small sample. The

post-transplantation sample was reduced partially because of the

increasingly difficult task of systematically following-up on each of the

recipients. The difficulties have arisen because of the larger numbers of

patients being seen for pre-transplantation evaluations as well as the

increasing number being transplanted and being hospitalized for shorter

periods of time.

The current study stemmed from the practical, clinical, and empirical

needs of continuing the work at the University of Florida in the psychological

factors of cardiac transplantation. From a practical standpoint, telephone

contact as a follow-up method allows for more convenient access to the

medical team than consistently presenting for additional clinical visits.

Clinically speaking, further information is needed about the long-term quality

of life of cardiac recipients. It is our impression that further contact

between the recipient and the hospital is desired by the patient for support,

but is currently under-served for the multiple-year surviving recipient. The

information gained from this study would provide additional data about the

long-term adjustment of transplantation recipients. However, the current

project goes beyond the work of other studies and attempted to predict the

long-term data with the personality and cognitive functioning pre-

transplantation. From an empirical standpoint, quality of life is increasingly

becoming an important outcome variable that may become a part of health

care planning and policy planning. Information about the pre-transplantation

predictors may allow for systematic interventions that could be tailored to

affect the quality of life of future recipients.


The present study was a longitudinal investigation of cardiac

transplantation recipients of Shands Hospital at the University of Florida.

The purpose of the present study was four-fold: 1) to provide descriptive

information about the prevalence of immunosuppressive symptomatology

and quality of life of cardiac transplantation recipients, 2) to test the

predictive validity of a pre-transplantation psychological assessment battery

on the recipient's perceived post-transplantation quality of life, 3) to test the

predictive validity of the pre-transplantation psychological assessment

battery on the perceived post-transplantation quality of life by spouse/proxy,

and 4) to assess the correspondence of the patient and proxy raters on

quality of life.

Structured telephone interviews were used to record the recipient's

and spouse/proxy's view of the recipient's quality of life. Quality of life was

defined as a multi-dimensional construct that included: a symptom report, a

satisfaction report, and a health report. The recipient and spouse/proxy

responses to these various instruments were then merged by the recipient's

medical record number to the recipient's pre-transplantation psychological

test results. The pre-transplantation psychological test results (measures of

intellectual and cognitive functioning, as well as personality), relevant

medical variables (e.g., ejection fraction), and demographic characteristics

served as independent variables in the prediction of two sets of perceptions

of post-transplantation quality of life, the recipient's and his/her

spouse/proxy's. This compilation of information allowed for the examination

of possible pre-transplantation "markers" that were associated with desired

post-transplantation quality of life. Correlations and planned comparisons

between the recipient and the spouse/proxy perceptions of the recipient's

quality of life were calculated to examine the differing perspectives on

quality of life.

The current study was designed to closely ally the science and

practice of psychology. The pre-transplantation test battery was composed

primarily of frequently used clinical instruments. In addition, due to the

relative dearth of longitudinal, predictive studies with cardiac transplantation

recipients many of the following hypotheses stemmed from clinical

experience and the research findings from other organ transplantation



1) Descriptive analyses of the indices of quality of life were expected to

show that the recipients have good quality of life as reported by both the

recipient and spouse/proxies.

2) It was expected that the pre-transplantation personality cluster groups

found by Robinson, Greene, and Geisser (1993) would be replicated in this

study and form four cluster groups.

3) The pre-transplantation personality cluster groups were expected to be

significantly different from one another on the post-transplantation quality of

life variables, while controlling for the relevant medical and/or demographic


If significance for personality cluster was secured, planned

comparisons were hypothesized as follows:

a) The High Distress group (elevations, in order, on the MMPI

subscales of depression, hypochondriasis, and hysteria (scales

2, 1, and 3)) will differ from the other 3 cluster groups and will

report the lowest life satisfaction and the lowest general health,

mental health, and physical functioning.

b) The Neurotic Triad group (elevations on the MMPI subscales of

hypochondriasis, hysteria, and depression (subscales 1, 3, and

2)) will report a high degree of disease symptoms similar to the

Subclinical V group. However, the Neurotic Triad group will

have better general health, mental health, physical functioning,

and satisfaction than the High Distress group.

c) The Subclinical V group (elevations on MMPI subscales of

hysteria, hypochondriasis, and depression (subscales 3, 1, and

2)) will report a high degree of disease symptoms and the low

general health. However, this group will have similar life

satisfaction, mental health, and physical functioning to the

Neurotic Triad group.

d) The Normal group (near elevation on the MMPI Mania subscale

(Scale 9) is expected to report the highest physical functioning,

mental health, and general health.

4) Similar predictions were made for the proxy ratings of the patient's

quality of life.

5) In regard to patient-proxy correspondence on the quality of life indices, it

was expected that all measures would be significantly related. However,

significant differences were hypothesized on the measures regarding role

limitations due to physical and mental problems. It was expected that proxy

raters would rate the recipient as significantly more limited than the recipient

rates himself\herself. In addition, significant differences were expected on

Mental Health, such that recipients would rate their functioning higher than

spouse/proxies would rate the recipient's functioning.



The sample included 83 cardiac transplantation recipients and 67

spouses/significant others from Shands Hospital at the University of Florida.

The majority of the recipients were male (84%), Caucasian (93%), and

married (83%). The mean age of the recipient sample was 55.9 years (sd =

8.8). The most common diagnosis for the recipients was ischemic

cardiomyopathy (65%) and idiopathic/viral cardiomyopathy (31%). Of the

spouses/proxies, 72% were female and 75% were spouses. The mean age

of the spouse/proxies was 54.7 years. Table 2 provides sample frequencies

on demographic variables. Additional descriptives regarding the medical

variables are presented in the Results section.

Subjects participated on a voluntarily basis and received no

compensation for their involvement. Informed consent procedures as

reviewed by the Institutional Review Board (IRB) at the University of Florida

Health Science Center were utilized. The inclusion criteria for participation in

this study were that all subjects were at least 3 months post-transplantation

and had completed a routine pre-transplantation psychological assessment

independent from their participation in this study. Attempts were made to

Table 2
Demographic Data



Variable n % n %




Marital Status

Employment Status

Education in Years
12 (H.S. Diploma)


70 84.3
13 15.7




7 8.6
32 38.6
29 34.9
13 15.6
2 2.4

53 65.4

7 10.4
60 89.6




9 13.4
29 43.3
22 32.8
6 11.1
1 1.4



contact all of the potential 108 recipients by telephone. The return rates

were approximately 78% for the recipients and 81% for the proxy raters.

These return rates compare favorably to studies with similar methodologies

(e.g., Magaziner et al., 1988: 49% for patients and 86% for proxy raters).

The majority of missing subjects (n = 12) were unable to be reached after

repeated attempts. Other reasons for not participating included: refusal to

participate (n = 6), death during study time period (n = 3), no valid phone

number (n = 2), and physically unable to participate due to hearing loss (n

= 1). The participation of spouse/proxies was solicited only after securing

data and permission from the recipient. Reasons for missing data in the

spouse/proxy group included: single/no proxy (n = 9), ineligible due to

assisting recipient in responding ( = 2), spouse/proxy deceased (n = 2),

refusal (n = 2), and unable to reach after repeated attempts (n = 1).


Following project approval by the Institutional Review Board, a master

list of cardiac transplantation recipients was obtained from the Division of

Cardiothoracic Surgery at the University of Florida/Shands Hospital.

Potential subjects were cross-checked by clinical research assistants in the

Psychology Clinic to ensure that psychological test results from the

recipient's pre-transplantation evaluation were available. After this

confirmation, letters from the principal investigator and the Director of the

Cardiac Transplantation Program were sent to the cardiac transplant

recipients describing the rationale for the study and seeking the participation

of the recipient and his/her spouse/proxy. Subjects were then contacted by

phone and semi-structured quality of life interviews were conducted by

research assistants. Data sheets containing the respondents' ratings were

completed by the interviewer during the phone interview. Confidentiality

was maintained by using only the patient's project subject number on the

data sheets in the data set. Subject numbers were later linked to medical

record number for the purpose of merging pre-transplantation data. All other

identifying information was deleted.

Telephone interviews consisted of two interviews, one interview

focused on the perceptions of the transplant recipient and one focused on

the perceptions of the recipient's spouse/proxy. The interviews were

performed consecutively with recipient typically responding to interview

questions first. In most cases, the interviews were performed on the same

day, during the same phone call. The interviews were conducted primarily

by trained undergraduate research assistants and lasted 30 minutes per

respondent. The training for the interviewers consisted of approximately 20

hours of didactic presentations of significant issues in cardiac

transplantation, case presentations, literature review, and role playing.


Pre-transplantation assessment battery. Intellectual screening was

performed using the Vocabulary and Block Design subtests of the Wechsler

Adult Intelligence Scale-Revised (WAIS-R). The raw scores obtained were

converted to Scaled Scores in which a Scaled Score of 10 is average and

the standard deviation is 3. Previous research has found that these two

subscales have the highest correlations with the Full Scale I.Q. (L = .89;

Ryan, Larsen, & Prifiteria, 1983). Vocabulary and Block Design are thought

to tap the abilities of verbal comprehension and perceptual organization,

respectively. Evidence exists that these scales can be summed to form an

extrapolated intelligence quotient (Silverstein, 1982).

The Wechsler Memory Scale (WMS) was given to assess the various

aspects of memory functioning. The WMS consists of seven subtests

including: Personal and Current Information (PI) and Orientation (OR) (e.g.,

questions related to ability to recall current information about patient's age,

location, and government officials), Mental Control (e.g., counting by 3's),

Logical Memory (e.g., memory of a story), Memory Span (e.g., digit recall

forwards and backwards), Visual Reproduction (e.g., memory for figures),

Associate Learning (e.g., memory for paired associates). The patient's raw

scores on these tests are summed and an age correction factor is added to

obtain a summary score called the Memory Quotient (MQ). Spreen and

Strauss (1991) have summarized the criticisms of the WMS. They list the

following limitations: 1) inadequate norms, 2) the MQ score does not

differentiate between the various facets of memory, 3) over-reliance on

immediate recall and no long-term memory sampling, 4) lack of control for

visuoperceptive and visuomotor abilities in so= called tests of visual

memory, 5) imprecise scoring criteria, 6) overemphasis on verbal tasks, 7)

orientation and mental control not genuine measures of memory. Despite

these criticisms, the WMS has been one of the most used measures of

memory in clinical practice.

Personality was assessed using the Minnesota Multiphasic Personality

Inventory (MMPI). The MMPI allows for the multi-dimensional assessment

of personality with related sampling of response-style distortions. The MMPI

was empirically derived which refers to the ability of the items to

discriminate between criterion groups such as normals and psychiatrically

diagnosed persons. In the current study, the 10 clinical scales, the four

validity scales, and four supplementary scales were used. The scales are as

follows: Scale 1 = Hypochondriasis, Scale 2 = Depression, Scale 3 =

Hysteria, Scale 4 = Psychopathic Deviate, Scale 5 =

Masculinity/Femininity, Scale 6 = Paranoia, Scale 7 = Psychasthenia, Scale

8 = Schizophrenia, Scale 9 = Mania, Scale 10 = Social Introversion. The

validity scales include: L = Lie Scale, F = Infrequency scale, K = Correction

scale, and ? = Question scale. The MMPI has been widely used despite

some limitations. Green (1985) reviewed the psychometric information

related to the MMPI and found that the reliabilities were generally acceptable

with 3-4 day test-retest reliability ranging from .56 to .88, with the majority

in the low .80s. Dahlstrom, Welsh, and Dahlstrom (1972) report extensively

on the MMPI's validity information.

Negative affect was assessed by the State-Trait Anxiety Inventory

(STAI; (Spielberger, Gorsuch, & Lushene, 1970), the State-Trait Anger

Expression Inventory (STAXI; Spielberger, 1984), and the Beck Depression

Inventory (BDI; Beck & Beamesderfer, 1974). Each of these measures are

among the most widely used to measure their respective constructs.

The STAI is a 40-item questionnaire developed to measure two

distinct components of the concept of anxiety. State anxiety is defined as

anxiety which a person experiences in response to certain, specific

conditions. In contrast, trait anxiety refers to a relatively enduring

personality characteristic, specifically, anxiety proneness. The internal

reliability of both the state and trait anxiety has been to shown to be

uniformly high across samples of adults ranging from .89 to .96. Test-retest

validity for the trait anxiety scale for college students over 104 days was

.73 for males and .77 for females. Concurrent validity between the STAI

and the IPAT Anxiety Scale and the Taylor Manifest Anxiety Scale ranged

from .83 to .73 (Spielberger, 1984).

The STAXI is a 44-item, self-report questionnaire designed to measure

the experience, control, and mode of expression of anger (Spielberger,

1988). Our analyses focused only on the state and trait anger subscales.

The State Anger subscale measures the intensity of angry feelings at a

particular time. The Trait Anger subscale provides a measure of individual

differences in the disposition to experience anger. The STAXI has excellent

psychometric characteristics including test-retest reliability ranging from .71

to .93 in college populations and concurrent validity with other hostility


The BDI is a 21-item questionnaire that measures the intensity of

depression-related phenomena such as mood, pessimism, sense of failure

guilt, suicidal ideation, etc. Beck, Steer, and Garbin (1988) meta-analyzed

the BDI's internal consistency and found a coefficient alpha of .86 for

psychiatric patients and .81 for non-psychiatric subjects. Concurrent validity

studies yielded a mean correlation with the Hamilton Psychiatric Rating Scale

for Depression of .72.

The psychological assessment battery, is a part of a comprehensive

medical evaluation performed on each transplantation candidate. The

different medical indices that were selected for this study are thought to be

sensitive to the various aspects of the cardiac transplantation medical

presentation. Many of these parameters are particular foci of the

transplantation review board decisions. For example, ejection fraction, the

global index of the extent of ventricular fiber shortening, is a good measure

of the ability of the heart to perform its duty of circulating oxygenated blood

to the body. Disease duration was determined by an archival record review

that found the length of time between the patient's recollection of first

cardiac event and eventual transplantation as measured by months.

Diagnostic type was restricted to three different diagnoses including:

ischemic and viral cardiomyopathy and congenital disorders. Concomitant

disease information was obtained from both medical record review and

interview to assess other illnesses that may be contributing to reductions in

the recipient's quality of life. Surgical and hospital-related complications

(e.g., infections) were assessed using the length of hospital stay between

transplantation and first discharge. Finally, time since transplantation was

computed from the number of months between transplantation and the

quality of life interview.

Post-transplantation Assessment Battery. Quality of life was

operationalized in our study as a multi-dimensional construct that included:

symptom report, satisfaction report, and health reports. The questionnaire

items are presented in the Appendix.

Symptom report. Because cardiac transplantation is expected to

reduce cardiac symptomatology, symptom report was operationalized by the

Transplant Symptom Frequency and Distress Scale that was designed for

use with cardiac transplantation patients taking immunosuppression drugs

(Lough et al., 1987). The scale samples responses to the frequency of 29

symptoms using a 0-4 rating (0 = never have symptom and 4 = always

have symptom). In addition, for each symptom that the recipient or proxy

acknowledges, they were asked, "Is this a problem for you?" It also is

important to note that this question is the only question that was asked

differently to the recipient and to the proxy/spouse. Specifically, we were

interested in what symptoms were problematic for the recipient but not the

spouse and vice versa.

The Transplant Frequency Scale demonstrated an alpha coefficient of

.70 in previous work by Lough et al. (1987). This scale resulted in a

Symptom Frequency score and a Problem Total score that were used in all

analyses. The Symptom Frequency index was formed by computing the

subject's reports of symptoms that occur "sometimes, often, or always."

Symptoms that occur with this frequency were given a value of "1" and

symptoms that occur "never or rarely" were given a "0" for that symptom.

Satisfaction report. The Satisfaction with Life Scale (SWLS)

developed by Diener et al. (1985) is 5-item scale designed to assess

satisfaction with the respondent's life as a whole. The items of the scale

are global rather than specific in nature which allows respondents to weight

domains of their lives in terms of their own values. Pavot and Diener (1993)

detail the psychometric validation of the instrument. Diener et al. (1985)

reported a coefficient alpha of .87 for the scale, and a two-month test-retest

stability coefficient of .82. In a four-year test-retest interval, the SWLS had

a .54 reliability coefficient (Diener, Magnus, & Fujita, 1991). The construct

validity of the SWLS has also been examined in relation to both self-report

and external criteria of measures of subjective well being and life

satisfaction. Pavot and Diener (1993) cite numerous articles that have

generally found that the SWLS is negatively correlated with measures of

distress and positively correlated with measures of positive affect. The

correlations between the SWLS and interviewer ratings have been positive,

ranging from .43 to .66. Additional research found that personal informant

ratings ranged from .28 to .58.

Health Report. The recipient's health-related quality of life was

assessed by the Short Form-36 (SF-36) that was both rationally and

empirically developed as part of the Medical Outcomes Study (Ware et al.,

1994). The SF-36 assesses 8 domains of health functioning including:

physical functioning, role limitations due to physical problems, role

limitations due to emotional problems, social functioning, bodily pain, mental

health, vitality, and general health perceptions. Descriptions of each of the

health domains and number of items are included below.

1) Physical functioning (10 items) refers to the extent that health

limits physical activities such as self-care, walking, climbing stairs, bending,

lifting, and moderate to vigorous activities. The physical functioning

measure is essentially equivalent to a functional status measure.

2) Role functioning due to physical problems (4 items) measures the

extent to which physical health interferes with work or other daily activities,

including such things as accomplishing less than desired or limitations in

type of activities.

3) Role limitations due to emotional problems (3 items) assesses the

extent to which emotional problems interfere with work or other daily

activities, including decreased productivity or quality of time spent on


4) Social functioning (2 items) refers to the extent to which physical

health or emotional problems interferes with normal social activities.

5) Bodily pain (2 items) measures the intensity of pain and effect of

pain on normal work.

6) Mental health (5 items) assesses general mental health including

depression, anxiety, behavioral-emotional control, and positive affect.

7) Vitality (4 items) samples the extent of feelings of energy versus

feelings of fatigue.

8) General health perceptions (5 items) measures personal

evaluation of health, health outlook, and perceived resiliency to illness. A

final single-item question examines the perceived changes in health from one

year ago in comparison to current health.

Despite having only a few items on each scale, the internal reliabilities

of the scales range from .77 (Social functioning) to .92 (Physical

functioning) in a sample of 3, 053 adults (Stewart, Sherbourne, Hays, Wells,

Nelson, Kamberg et al., 1992). The items and scales of the SF-36 are

scored so that a higher score indicates a better health state.

The SF-36 manual (Ware et al., 1994) provides a prescribed strategy

for handling missing data with SF-36 items. The authors strongly suggest

that missing items should be replaced by the average score on the other

items of the scale, provided a majority of items were answered on the scale.

This method was used manually to replace individual items in the event of a

missing item.


Psychological variables. The various indices of pre-transplantation

psychological functioning were examined in two ways. First, descriptive

statistics of the entire pre-transplantation assessment sample (n = 298)

were examined for each of the psychological variables of interest. Second,

the indices of pre-transplantation psychological functioning were examined

in the recipient group surveyed in our study. The descriptives for these two

samples are available in Table 3.

The group means were tested to determine if there were any

significant differences between the candidate and recipient groups. Families

of comparisons were created and Bonferroni correction procedures were

implemented. The families included cognitive (3 comparisons), affective (5

comparisons), and personality functioning (13 comparisons). Results

indicated significant differences on the Vocabulary subscale of the WAIS-R

[t (1, 165) = -2.47, 2 < .01] and on the Memory Quotient [t (1, 130) = -

2.74, p < .011, such that recipients were found to have significantly greater

scores on both measures. Results from tests of the affective measures

found that Trait Anxiety was significantly different [t (1, 83) = 2.41, a <

.01], such that the recipient group reported less Trait Anxiety at the pre-

transplantation assessment. In addition, the groups differed on State Anger

Table 3

Pre-Transplantation Intellectual and Cognitive Functioning Means



Variable n M sn M S

Cognitive Functioning

Vocabulary* 202 9.53 3.03 79 10.43 2.59

Block Design 195 7.75 2.68 78 8.29 2.48

Memory Quotient* 208 105.13 18.18 70 111.52 16.41

Affective Functioning

State Anger* 117 11.89 3.77 48 10.83 1.84

Trait Anger 117 16.12 4.59 48 15.31 4.47

State Anxiety 119 39.02? 10.80 44 36.52 13.12

Trait Anxiety* 117 37.52 9.46 43 33.77 8.44

Depression 205 11.43 6.90 70 11.14 6.38

* < .01

It (1, 158) = 2.41, a < .01] with the recipients reporting significantly less

anger. In the personality domain, only the F scale of the MMPI was

significantly different [1 (1, 224) = 3.22, a < .0011, such that the

recipients scored significantly lower. Table 4 provides the means tested in

these analyses.

Medical variables. The results from descriptive analyses with the

medical variables found that the sample at the pre-transplantation evaluation

was demonstrating severe cardiac dysfunction. The average ejection

fraction for the sample was 17.03 (s = 5.61) and the average disease

duration ranged from 3 months to 660 months (M = 133.8 months, s =

114.5). Our index of post-surgical complications was the number of

hospitalization days following transplantation. The post-hospitalization

period was also variable ranging from 10 to 53 days with an average 20.5

days (s = 8.1). Finally, the mean time since transplantation to the date of

our contact was 36 months (s = 25.3) with a range of 4-100 months.

Post-TransDlantation Quality of Life Measures

Symptom report. According to the recipients, they experienced an

average of approximately 10 symptoms. The most common symptom

reported was fatigue with 71% of the recipients reporting that they were at

least sometimes fatigued. Other symptoms that were most prevalent

included: excessive hair growth (67.5%), poor concentration (49.4%),

changed bodily appearance (49.4%), bruises (48.8%), fragile skin (48.2%),

mood swings (46.3%), and lack of sleep (45.8%). Figure 1 displays the

complete set of symptom reports. Consistent with the most common

symptom, the most problematic symptom was fatigue (40.2%). However,

the symptom that was reported by the recipients as the second most

problematic was difficulty concentrating (29%). Other problematic

symptoms endorsed by the recipients included: fragile skin (26.5%),

impotence (25.7%), pain (25.3%), changed bodily appearance (24.1%), and

swollen ankles (22.9%). Depression was reported by 33.7% and rated as

problematic by 20.5% of the sample. Figure 2 displays the prevalence of

perceived problems with each symptom.

A slightly different picture of the most common symptoms

experienced by the recipients was provided by the spouse/proxies. The

spouse/proxies indicated that excessive hair growth was the symptom

experienced at least some of the time by the largest number of recipients

(74.6%). Fatigue was the second most common symptom as viewed by the

spouse/proxies (67.2%), followed by bruises (58.2%), mood swings

(56.7%), and lack of sleep (56.3%). Depression was reported as a common

symptom for 40.3% of the recipients, as perceived by their spouse/proxies.

Figure 3 displays the complete set of symptom reports by the


The Problem Total from the spouse/proxy ratings involved sampling

symptoms that are a source of problems for the spouse/proxy. According to

the spouse/proxies, the recipient symptom most problematic for them was

Table 4

MMPI Means of Candidates and Recipients



Variable M M

Personality Functioning (MMPI)








































n = 216. bn = 82.

* < .01















c 0




C =




a) 0

a) 4)

CL c

cN(mo q Lo DN o



. "

C 0

U uj


C 0o
2m'( > o
ti- CL 0 CL





I .


m 0
I 0


0 a




.. I
> 11

0 0 0 0
O 4 C4






















r ur

Co o

a ao> emO o
LU gO 1c .

co ) r OL c


c0 GO^O v NMC4 0 C

0 QAc
0 c


S5 a e-- # r a

E >n ii in os
0 0 COO) O
. > OrECmCOw
U. 03 T-N O^-lO


ip I













0 0-
L2 4- 0)

OC 1 | 0
.o* 0C 00 .E-
am0 0 o. 0 0-4
0EEa 4t 00o M
3 4 = E C 2 3W
QZI -2mcCL AQ0



E g > >M Mo'.oL0

.. a II II C I

SA V N C N 0 M V I I I

ii | il:i~l~srgis



0 0 0 0



O m c

jte 0
0 jr0 C*-C

0) O0, 00
0 000 w 0


0- a c a w
o o -c a .. o cS c e

a a) cc 0 4 ,
0 : 0 0 2 S ( 0 > 0 a
E e omE 1oomO3> oL

o aE

a) o

%.. M -t 0L
E C0 heg N M q 0 W N MM
*~ 5 o(o l l 2 t*c S "2'5
^-c >Q u a Uuj< t 0C
^ ^ u 11 11 1 1 n n n i "
.1' ^ "" "o -N w^ i
u. i-Mc>^to or~oo i--i--T











__ o


O 0 Oto o

o) CO_ __-* q

Cv, cv c I _

mood swings (31.5%), followed by impotence (24.4%), changed bodily

appearance (24.2%), and decreased sexual interest (22.9%). Depression

was noted as problematic by 20.4% of the spouses. Figure 4 provides a

graphic representation of both the recipient and spouse/proxy symptom


Satisfaction report. The scores on the SWLS obtained from recipients

indicated that they feel a high degree of satisfaction with life (M = 23.67, s

= 7.35) that is commensurate with a normative sample of healthy, older

adults (M = 25.2, s = 6.7, n = 69; Pavot & Diener, 1993). Similarly, the

proxy/spouses reported that they believed that the recipients had a high

degree of Life Satisfaction (M = 23.94, s = 7.19).

Health report. The SF-36 scores obtained from the recipients

indicated that their reported Physical Functioning, Mental Health, and

General Health are within normal limits. Compared to a general United

States sample of men and women ages 55-65 (Ware, 1993), these three

health indices would place the study sample between the 25th and 50th

percentile. Similar scores were found in examining the spouse/proxy reports

such that Physical Functioning, Mental Health, and General Health scores

were within normal limits, as compared to a general U.S. same age sample.

Table 5 provides a full report of both recipient and proxy/spouse scores for

each of the SF-36 subscales.

Correlations for Inclusion/Exclusion

Pearson's correlational coefficients were computed to assess the

relationships between the full array of potential independent variables

representing the medical, demographic, and psychological domains and each

of the quality of life variables from the recipient and spouse/proxy raters.

Because of the large number of correlations computed, no attempt to

systematically interpret each of the correlations was attempted. Instead, the

correlations were used to reduce the potential number of covariates by using

an inclusion criteria that involved a correlational coefficient with a

significance level of D <.01. The significant correlations will be reported


The demographic correlations were somewhat tenuous because the

categories had to be collapsed to form cells with a sufficient size to perform

meaningful correlations with the outcome measures. For example, race was

collapsed into Caucasian and non-Caucasian. Similarly, marital status was

collapsed into married and non-married. Finally, employment was converted

to working (full or part-time) and non-working (retired, unemployed,

medically disabled). Despite our attempts to sufficiently test for potential

relationships, no significant correlations (Pearsons or biserial) were found

between the demographic variables and any of the outcome measures from

either the recipient or spouse/proxy raters.

The medical variables including days of post-transplantation

hospitalization, ejection fraction, time since transplantation, diagnostic type

Table 5

Quality of Life Grand Means



Variable n M a n M s

Symptom Report

Problem Total 77

Frequency 77

Satisfaction Report (SWLS)

Life 83

Health Report (SF-36)

Physical Function 83

Role-Physical 83

Role-Emotional 82

Social 83

Mental Health 83

Vitality 83

General Health 83

4.57 4.22

9.78 5.22

23.67 7.35








50 3.98 4.04

42 10.74 5.17

66 23.94 7.19






















(ischemic cardiomyopathy and viral/idiopathic cardiomyopathy), and disease

duration fared somewhat better in establishing relatively strong correlations

with the outcome measures. For example, days of post-transplantation

hospitalization was significantly correlated with both the recipient and

spouse/proxy rater's scores for General Health (r = -.31, 2 < .01 and r = -

.28, D < .01, respectively). Time since transplantation was also correlated

with the recipient and spouse/proxy rater's scores for General Health scores

( = -.27, p < .01; r = -.35, p < .01), respectively. Time since

transplantation was also found to be related to the spouse/proxy reports of

Physical Functioning (r = -.34, < .01).

The psychological assessment battery was also subjected to

correlational analyses. The variables included in this analysis were: trait

anger, trait anxiety, the Wechsler Memory Scale subscales, and the WAIS-R

subscales of Vocabulary and Block Design. Trait Anxiety was also found to

be correlated with both Symptom Frequency (r = .49, 2 < .01) and

negatively correlated with Mental Health (r = -.49, p < .01). Mental health

was also highly correlated with the Beck Depression Inventory (- = -.31, a

< .01) suggesting that the mental health index taps mixed emotional

features. In sum, the variables that emerge from the correlational analyses

that will be included as covariates in the MANCOVA analyses include: the

medical variables of time since transplantation and post-transplantation

hospitalization time and no demographic variables.

Cluster Analyses

The scores from the MMPI for the full sample of cardiac

transplantation candidates were subjected to cluster analyses for the

purpose of defining homogeneous groups of candidates with similar scores

across the subscales of the MMPI. This strategy was intended to replicate

the cardiac transplantation candidate profiles already demonstrated by

Robinson, Greene, and Geisser (1993) with a subsample (n = 160) of the

current population (n = 298). The Ward method of clustering was utilized

because it attempts to minimize the variance within groups and maximize

the between group variance. Four groups were specified. The results from

this analyses indicated that the larger sample provided generally similar

profiles to Robinson et al. (1993) with some notable differences on specific

subscales. Cluster 1 (n = 85) was identified as the "Subclinical V-type"

such that the highest scale scores were on scale 1 and 3 and scale 2 slightly

lower forming a "V" shape when plotted. Cluster 2 (n = 85) was identified

as the "Normal" group with no significant elevations on any MMPI subscale.

This cluster differed slightly from the Robinson et al. (1993) "Normal" group

in that scale 9 was significantly lower than their finding in which this group

had a near elevation for scale 9. Cluster 3 (n = 110) was identified as the

"Neurotic Triad" group. This cluster group is marked by significant clinical

elevations on scales 1, 2, and 3, suggesting affective distress. Finally,

Cluster 4 (n = 17) was given a new label, the "Distressed/Confused" group,

with significant elevations on all of the clinical scales except scale 5 and 10

and the F scale. This group was not described as the "High Distress" group

consistent with Robinson et al. (1993) labels because there were more

clinical elevations in this group after our analysis. Of particular importance

was a clinical elevation on scale 8 suggesting mental confusion or unusual

sensory experiences consistent with psychotic symptomatology and/or

functioning. Figure 5 provides a graphic representation of these cluster


With cluster group membership having been assigned to all

transplantation candidates, we sought to examine the frequencies of each

cluster among our sample of recipients. This analysis showed that the

frequencies for 3 of 4 of the groups was consistent with the larger sample

as shown here: Cluster 1 (n = 24; 29.3%), Cluster 2 (n = 22; 26.8%), and

Cluster 3 (Q = 35; 42.7%). However, Cluster 4 clearly was not well-

represented (n = 1; 1.2%). Our medical record review of the full roster of

recipients including those recipients who did not participate in our study

showed that only 1 recipient who was classified in Cluster 4 is still being

followed by the Department of Cardiology. The minimal sample size

necessary for additional statistical analyses necessitated the omission of

Cluster 4 from future analyses. The resulting clusters are shown in Figure 6.

A primary purpose of this study was to examine possible differences

between cluster groups on quality of life variables. Multivariate Analyses of

Covariance (MANCOVA) were used to test for differences between groups.

The covariates (time since transplantation and post-transplantation

hospitalization time) were used due to the significant correlations that were

a g


^" "
0 .I

> .c

. CC >

) >-0 "-
10, 2


o 0 O
)..C 0 0

to Q
aIn w e

CL 0


0 0 0 0 0
co r- CD 1,0

found in the analyses reported above. Two families of dependent variables

were used for these analyses: recipient ratings and spouse/proxy ratings.

Specifically, 6 tests of cluster differences were performed for both families

using the Symptom Frequency, Problem Total, Satisfaction Report, Physical

Functioning, Mental Health, and General Health as the dependent variables.

The means for each of the cluster groups on the quality of life measures are

presented in Table 6.

The family of recipient ratings were examined first. For the Symptom

Frequency measure, MANCOVA procedures found no significant differences

between cluster groups [E (2, 68) = 1.47, P > .05]. On the second

analysis for the Problem Total variable, no significant differences between

cluster groups were found [E (2, 68) = .87, > .051. The Satisfaction

Report was tested next for cluster group differences. However, no

significant differences were found [F (2, 74) = .28, p. > .05].

The other three MANCOVA procedures using the Health Report

found no significant differences for the clusters groups on Physical

Functioning [E (2, 74) = .93, p > .05], Mental Health .E (2, 74) = .29, p

> .05], or General Health IF (2, 74) = .23, a > .05]. However, post-

transplantation hospitalization time (p < .05) was a significant covariate for

Mental Health, such that increased post-transplantation hospitalization was

associated with decreased mental health functioning. The covariates in the

other comparisons were found to be non-significant. Taken together, these

analyses suggest that MMPI cluster membership in one of three groups


oo c0
>c.- o

00 *a)U a

> 111 11 II
= 0 CL m c

25 uo o. IL no. US
_3L he-i UL e m a
u~ Olo~r.
EC ~ r
a)u )Sr( ~

0 S I *I ~Q | -


a. i


I u



0 0 0 0 0 0 0
0) cO I -P- o

O 0o i/ ; CO 0
f) l( fl rl i -

Table 6

Mean Recipient Quality of Life Ratings by Cluster Grouo

Cluster 2b

Cluster 3*

Variable M A M A M s

Svmotom Report

Problem Total 4.39 4.23 3.81 3.72 5.26 4.63

Frequency 8.74 5.80 9.52 4.62 10.84 5.29

Satisfaction Report (SWLS)

Life 24.79 7.70 23.23 7.39 23.83 6.89

Health Report (SF-36)

Physical Function 72.50 23.50 76.36 21.45 69.57 22.25

Mental Health 75.67 14.59 71.27 18.99 72.00 23.24

General Health 61.54 23.06 63.00 28.90 60.51 23.27

"n = 24. bn = 22. "n = 35.

Cluster 1"

tested is unrelated to post-transplantation functioning. However, increased

post-transplantation hospitalization time is associated with increased

symptom reporting and decreased mental health.

The same order of analyses was conducted using the spouse/proxy

ratings on the quality of life measures. The results from MANCOVAs with

the Symptom Report as the dependent variable were consistent with the

recipients', such that no significant differences were found between cluster

groups [L (2, 60) = .97, p > .05]. Results from tests using the Problem

Total as the dependent variable found no significant differences for the

cluster groups E (2, 34) = .04, p > .05]. On the Satisfaction report, the

cluster groups were found to be not significantly different IE (2, 59) = .07,

p > .05].

The final three MANCOVA procedures using the spouse/proxy Health

Report found no significant differences for the clusters groups on Physical

Functioning [F (2, 60) = .47, 2 > .05], Mental Health [E (2, 59) = .33, p

> 05]., or General Health [F (2, 60) = .83, p > .05]. However, time since

transplantation (2 < .05) was a significant covariate for Physical

Functioning, such that increased time since transplantation was associated

with decreased physical functioning. Analyses using these covariates in

alternative comparisons were found to be non-significant. In sum, the

findings from these comparisons provide no evidence for MMPI cluster

membership differences on post-transplantation functioning for the three

groups tested using either recipient or spouse/proxy raters. The means for

each of the cluster groups on the spouse/proxy ratings of quality of life

measures are available in Table 7.

Regression Analyses

A second primary purpose of this study was to examine the possibility

of variables serving as "markers" for prediction of quality of life. Stepwise

regression analyses were utilized to examine this possibility. Using the 6

dependent variables of quality of life, the variables of the medical domain

(e.g., time since transplantation, ejection fraction, duration of disease, and

post-transplantation hospitalization time) and the psychological domains (e.g.

Trait Anxiety, Trait Anger, and Depression scores) were entered to

determine which variables accounted for unique variance. Each analysis

was performed in an identical, stepwise fashion with the variable accounting

for the most variance being entered first, followed by additional significant

variables in descending order of variance accounted for by the independent

variable. Missing variables were handled pairwise, such that all possible

variables with valid values on the variables being compared were included in

the analysis. The stepwise procedure was selected because no order of

entry was predicted. This procedure was thought to maximize the

possibility of finding a predictor in the least number of analyses.

Symptom Frequency was the first dependent variable used in the

regression analyses. Trait Anxiety was the first variable entered into the

equation IE (1, 37) = 11.75, 1 < .01], accounting for 24% of the variance.

No other predictor variables entered the regression equation. The next

Table 7

Mean Soouse/Proxv Quality of Life Ratings by Cluster Group

Cluster 2b

Cluster 3c

Variable M s M s M s

Svmotom Report

Problem Total 3.67 4.17 4.00 3.74 4.30 4.40

Frequency 9.00 5.32 11.00 5.94 11.40 4.83

Satisfaction Report (SWLS)

Life 23.68 7.17 24.33 5.97 24.48 7.99

Health Reoort (SF-36)

Physical Function 60.23 29.09 71.25 30.30 69.44 25.88

Mental Health 66.29 20.81 65.00 17.68 70.52 23.93

General Health 57.95 33.87 64.94 21.48 54.26 29.60

a = 22. bn = 22. an = 27.

Cluster 1"

dependent variable examined was the Problem Total score. Once again, only

Trait Anxiety entered the equation IE (1, 37) = 5.64, < .051, accounting

for 13% of the variance. The third regression analysis involved the

prediction of the Satisfaction report. However, no significant predictors

were found.

The final three regression equations with the recipient ratings involved

the Health Report variables. On the Physical Functioning measure, no

significant predictors entered the model. Mental Health was examined next

and Trait Anxiety entered the model lE (1, 37) = 11.47, 2 < .01],

accounting for 24% of the variance. No other variables entered the model.

Finally, General Health was examined using regression analysis. Post-

transplantation hospitalization entered the model (E (1, 37) = 4.34, D <

.05] accounting for approximately 11 % of the variance. No additional

variables entered the model for General Health.

Spouse/Proxv Regression Analyses

The spouse/proxy ratings were examined in an identical fashion to the

recipients' ratings using regression analyses. The medical variables of time

since transplantation, ejection fraction, duration of disease, and post-

transplantation hospitalization and the psychological variables of Trait

Anxiety, Trait Anger, and Depression were used as the independent

variables. Symptom Frequency was used as the first dependent variable,

however, no significant predictors were found. Similarly, no significant

predictors were found for Problem Total. The Satisfaction report was the

third regression analysis performed. Trait Anxiety entered the model [E (1,

29) = 6.05, p < .05] and accounted for 17% of the variance in life

satisfaction. No other variables were found as significant predictors.

In the regression equations using Physical Functioning as the

dependent variable, no significant predictors were found. However, in

regression equations using the Mental Health score as the dependent

variable, Trait Anxiety entered the model [E (1, 30) = 4.93, p < .05],

accounting for 14% of the variance. No other significant predictors were

found. Finally, General Health was tested for possible predictor variables but

no significant predictors were found.

Comparisons of Quality of Life (Recipients vs. Spouses/Proxies)

Correlations. Pearson's correlational coefficients were computed

between recipients and their spouse/proxies for each of the quality of life

variables and all correlations were statistically significant at the p < .0001

level, except for the Problem Total. Recipients and spouse/proxies

corresponded highly on the frequency of symptoms (r = .71, p <.0001) but

the Problem Total was not significant (r = .20, p <.09). This discrepancy

was somewhat expected because the Problem Total score was the only set

of questions that asked for the spouse/proxies to report how the recipient

impacted them personally. The Satisfaction Report from the recipient and

spouse/proxies was consistent (I = .59, a <.0001), suggesting a high

degree of agreement between the recipient and spouse/proxy views of the

recipient's overall life satisfaction. Finally, the indices of the Health Report

were each highly consistent between recipient and spouse/proxies yielding

significant correlations in Physical Functioning (I = .57, a <.0001), Mental

Health (r = .54, a <.0001), and General Health (f = .51, a <.0001).

Planned Comparisons. T-tests were used to compare the mean

ratings on each of the quality of life indices for significant differences

between recipients and the spouse/proxies. A total of 6 comparisons were

made and Bonferonni corrections were made to control for family-wise error

rate. The means are slightly different for the recipients' reports in these

comparisons because only recipients with spouse/proxy raters were included

in the analysis. Results indicated that significant differences were found

between recipients and spouse/proxies in their views of the recipients'

mental health [t (1, 66) = 2.37, p < .011, such that the spouse/proxies

reported significantly more impairment in mental health functioning. Table 8

provides the means used in these comparisons. It is interesting to note that

there were no differences found in the hypothesized direction in either role

functioning related to physical health or role functioning related to mental

health. In addition, no differences were found on Problem Total, which may

have been expected given that the respondent's were asked to respond

whether the symptom was a problem specifically for himself/herself and not

the recipient.

Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E2FPH4Z3B_KRVXSH INGEST_TIME 2012-08-30T15:57:36Z PACKAGE AA00011786_00001