3: o
it, 3.
a. ,

.1. than.
49:53.74
«12 1.31 n
. w la

 

 

 

 

 

 

 

 

 

 

 

 

)u.

 

 

 

”4&6
LIBRARIES
MICHIGAN STATE UNIVERSITY

EAST LANSING, MICH 48824-1048

This is to certify that the
dissertation entitled

THE SYMPTOM EXPERIENCE OF PATIENTS WITH
END-STAGE RENAL DISEASE ON HEMODIALYSIS

presented by

ANITA MARIE JABLONSKI

has been accepted towards fulfillment
of the requirements for the

Ph.D. degree in Nursing

 

 

4446M; / 1/1/ C

Major/V frofessor’ § Signature

27m ’7”! 4M L/

7Date

MSU is an Affirmative Action/Equal Opportunity Institution

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/ClRC/DateDue.p65-p.15

 

THE SYMPTOM EXPERIENCE OF PATIENTS WITH END STAGE RENAL
DISEASE ON HEMODIALYSIS

By

Anita Marie J ablonski

A DISSERTATION

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

DOCTOR OF PHILOSOPHY
College of Nursing

2004

ABSTRACT

THE SYMPTOM EXPERIENCE OF PATIENTS WITH END STAGE RENAL
DISEASE ON HEMODIALYSIS

By

Anita Marie J ablonski

The primary purpose of this study was to document the symptom experience of
End-Stage Renal Disease (ESRD) patients on hemodialysis as the first step in
determining if there is a need for improved symptom management. To accomplish this
goal, an in-depth examination of the symptom experience associated with ESRD was
completed. A convenience sample consisted of 130 hemodialysis patients, ages 22 to 82
years, recruited from two in-center dialysis clinics. Structured interviews were conducted
during dialysis sessions using a demographics questionnaire, the Chemistry Abnormality
Score, the Hospital Anxiety and Depression Scale, an ll-item multidimensional symptom
scale and corresponding relief scale, and the Ferrans’ and Powers’ Quality of Life Index.
Patients reported a high number and variety of distressing symptoms, many of which
were not satisfactorily relieved. Furthermore, high levels of symptoms and low levels of
relief were linked to diminished QOL. These results indicate that further investigation of
the level of relief patients obtain fiom their symptoms as well as effective strategies to

improve symptom management is warranted.

ACKNOWLEDGEMENTS
Audrey G. Gift, R.N., Ph.D., FAAN, Dissertation Committee Chairperson, is especially
acknowledged for her support during the completion of this dissertation. Her endless

encouragement and guidance contributed to the success of this project.

The following Dissertation Committee members are also recognized for their generous
assistance:

Deborah Kashy, Ph.D.,

Linda Spence, RN, PhD.

Gwen Wyatt, RN, PhD.

A special thank you to my fellow student, Susan Dunn, for sharing this experience with

me.

Funding sources:

National Institutes of Health, National Institute of Nursing Research. National Research
Service Award Grant No. 1 F31 NR08463-01.

American Nephrology Nurses Association/Sigma Theta Tau International

Sigma Theta Tau; Alpha Psi Chapter, Michigan State University

Sigma Theta Tau; Kappa Epsilon Chapter, Grand Valley State University

iii

TABLE OF CONTENTS

LIST OF TABLES ................................................................................... v
LIST OF FIGURES ................................................................................. vi
CHAPTER 1
INTRODUCTION ........................................................................ 1
CHAPTER 2
REVIEW OF THE LITERATURE ................................................... 4
Theoretical model .................................................................. 10
The Symptom Experience of Hemodialysis Patients ......................... 16
CHAPTER 3
METHODOLOGY ...................................................................... 34
Research questions ................................................................ 34
Sample ............................................................................... 35
Measures ............................................................................ 39
Procedures ......................................................................... 56
Data analysis ....................................................................... 59
CHAPTER 4
RESULTS ................................................................................ 62
CHAPTER 5
DISCUSSION ........................................................................... 96
Limitations ........................................................................ 1 10
Implications for Nursing Research and Practice ............................ 112
APPENDICES ...................................................................................... 119
MEASURES ............................................................................ 120
Demographic Information ....................................................... 120
Chemistry Abnormality Score ................................................... 123
Symptoms and Level of Symptom Relief ..................................... 124
Hospital Anxiety and Depression Scale ......................................... 148
Ferrans’ and Powers’ Quality of Life Index - Dialysis Version ............ 151
REFERENCES .................................................................................... 156

iv

 

LIST OF TABLES

Table 1. Demographic Characteristics of Patients by Clinic and Total Sample ........ 37-38
Table 2. Major Study Variables and Their Measurement ................................. 41-42
Table 3. Physical and Mental Health Characteristics of the Sample .......................... 63
Table 4. Prevalence of Self-Reported Comorbid Conditions ................................. 64
Table 5. Prevalence of Symptoms Reported by ESRD Hemodialysis Patients ............ 67

Table 6. Correlations Among Situational/Demographic, Psychological,
and Physiological Influencing Factors ................................................. 69

Table 7. Summary of Regression Equations Identifying Predictors of
Symptoms ............................................................................. 71 -72

Table 8. Average Characteristics of Symptoms Reported by Hemodialysis
Patients ..................................................................................... 75

Table 9. Average Symptom Scores, Symptom Dimension Ratings, and
Level of Relief Scores for Each Symptom ........................................... 77

Table 10. Correlations Between Symptom Dimensions ....................................... 78

Table 11. Results of Principal Components Factor Analysis with Varimax

Rotation Including Factor Loadings.. . .. ........80
Table 12. Change in Individual Symptom Scores with Increasing Numbers

of Symptoms in the Cluster ........................................................... 82
Table 13. Strategies Used by Patients to Relieve Symptoms ............................. 85-86

Table 14. Correlations Among Trimmed Influencing Factors, Symptoms,
Level of Symptom Relief, and Quality of Life ..................................... 88

Table 15. Summary of Regression Equations Identifying Predictors of
Quality of Life ........................................................................... 91

LIST OF FIGURES

Figure 1. Traditional model of palliative care .................................................... 6
Figure 2. Integrated model of palliative care ...................................................... 6
Figure 3. Suggested model of palliative care for ESRD patients .............................. 6
Figure 4. Theory of Unpleasant Symptoms ...................................................... 1 1

Figure 5. Theoretical model of the symptom experience of
hemodialysis patients .................................................................... 13

Figure 6. Quality of life scores relative to the number of
symptoms in the cluster reported ...................................................... 83

Figure 7. Revised theoretical model ............................................................... 95

vi

Chapter 1
INTRODUCTION

According to the United States Renal Data System (USRDS) (2002), nearly
400,000 persons suffer from end-stage renal disease (ESRD). This condition requires
long-term dialysis or transplantation to sustain life because the kidneys have lost the
ability to remove metabolic waste products and maintain internal homeostasis. Due to the
lengthened lifespan made possible by modern science, as well as the escalating elderly
population, the number of patients who will require treatment for ESRD is expected to
rise (Reikes, 2000). In fact, the incidence of ESRD is increasing at a rate of 5% per year
and the prevalence has doubled since 1988 (Reikes, 2000). This grth in the ESRD
population represents a major public health problem that affects individuals and society.

Of major concern are the costs associated with ESRD. The illness and its
treatment necessitate lifestyle changes that are costly both economically and personally.
The fiscal survival of the family unit is often jeopardized by employment changes as well
as the substantial costs of care (Sloan, 1999). The stress of coping with a severe chronic
illness erodes the emotional stability of the patient and family. Physiological alterations
that accompany ESRD and hemodialysis give rise to additional comorbid conditions and
a variety of symptoms that exact a toll as well. Because ESRD impacts every aspect of
life, it threatens the ability of patients and their families to maintain an acceptable quality
of life (QOL) (Ferrans & Powers, 1993).

The costs to society are also an issue. Medicare subsidizes the treatment of over
90% of afflicted persons (Reikes, 2000). In 1991 the charge to the government for the

ESRD program was $5 billion (Rettig & Levinsky, 1991). Expenditures currently exceed

 

$12.5 billion, representing 5.8% of the entire Medicare budget (USRDS, 2002). In light
of this financial outlay, it is understandable that there is considerable interest in the
quality of the care ESRD patients receive.

This concern was first evident in 1987 when Congress enacted the Omnibus
Reconciliation Act (OBRA). The OBRA legislation mandated that the health care
services purchased by the government on behalf of ESRD beneficiaries be monitored for
quality (Rettig & Levinsky, 1991). Congress delegated this task to the Institute of
Medicine (IOM). The ensuing IOM study assessed a number of traditional outcomes
including mortality, infection rates, and laboratory values. Less conventional quality
indicators such as patient reported symptoms, well-being, and QOL were also examined
(Rettig & Levinsky, 1991). Inclusion of these outcomes signaled a growing emphasis on
how well the health care system meets the psychosocial as well as the physical needs of
patients living with chronic illnesses that require treatment over an extended period of
time.

In recent years, symptom relief and QOL have assumed greater emphasis as
outcomes of care. Especially in the cancer population, palliative care services that
specialize in symptom management have become a routine component of the treatment
plan. Patients in the later stages of other chronic illnesses such as congestive heart failure
and chronic lung disease have also benefited from these services. Patients who suffer
from ESRD are rarely recipients of palliative care despite the fact that their illness as well
as its treatment produces unpleasant symptoms.

Although some renal experts have voiced the need for palliative care in the ESRD

pOpulation, the call has gone unheeded by the general healthcare community (Moss & the

ESRD Workgroup, 2001; Poppel, Cohen, & Germain, 2001). However, the need for
palliative care has not been well documented empirically. The symptom experience
associated with ESRD has been described to a limited extent in the nephrology literature,
focusing mainly on individual symptoms and their severity or frequency. Equally lacking
is research that verifies the level of symptom relief achieved by ESRD patients and the
extent to which unrelieved symptoms influence QOL (Hoffart, 1995). Therefore, the
primary aim of this exploratory study was to generate a comprehensive description of the
symptom experience of ESRD patients, including the interplay among symptoms, level of
symptom relief, and QOL. Filling in the knowledge gaps surrounding symptom
experience is necessary to achieve the ultimate outcome of this investigation, namely, to
determine whether there is a need for improved symptom relief to enhance the QOL of

ESRD patients.

Chapter 2
REVIEW OF THE LITERATURE

Only 50 years ago, patients diagnosed with chronic, irreversible renal failure
faced certain death. Since that time, however, scientific innovations have resulted in
effective treatment for individuals who have lost kidney function. The development of
renal replacement therapy (RRT), including dialysis and transplantation, has made it
possible to prolong the lives of patients even after they progress to end-stage renal
disease (ESRD).

Of all forms of RT, hemodialysis is the treatment of choice for the majority of
ESRD patients. Approximately 87% of incident (i.e., at the time of first regular dialysis)
patients begin treatment with hemodialysis (United States Renal Data System [USRDS],
2002). Sixty-five percent of prevalent (i.e., known to be receiving treatment for a period
of time) dialysis patients less than 65 years of age undergo this form of therapy. The
proportion rises to 85% for prevalent dialysis patients over 65 (USRDS, 2002). Because
of the predominance of hemodialysis as the preferred treatment modality, this
investigation focused on this subgroup of the ESRD population.

Although hemodialysis is a life saving therapy, it is not without unpleasant side
effects. The underlying disease process and the presence of additional comorbid
conditions further add to the number and variety of symptoms experienced by
hemodialysis patients. Undocumented clinical observations suggest that many of these
symptoms are not adequately controlled and that healthcare providers’ symptom
management efforts have been inadequate or poorly informed (Poppel, Cohen, &

Germain, 2001).

It is proposed that the efforts to control unpleasant symptoms associated with
hemodialysis and ESRD would be enhanced by the use of palliative care strategies. This
recommendation is strongly endorsed in a recent report compiled by the ESRD
Workgroup, a subgroup of The Robert Wood Johnson Foundation program Promoting
Excellence in End-of-Life Care (Moss & the ESRD Workgroup, 2002). Following a
comprehensive review of the current state-of-care, the subgroup concluded that there is
an urgent need for palliative care in the ESRD population.

Although considered to be a critical component of comprehensive care, ESRD
patients are rarely the recipients of palliative care services. The reason lies in the
traditional view of palliative care as a service limited to cancer patients who opt to forgo
aggressive, curative treatment. In this paradigm, curative care and palliative care are
viewed as two distinct entities, such that active treatment ends when palliative care
begins (see Figure 1). Because the majority of ESRD patients continue active treatment,
palliative care is not considered a routine part of care.

Active treatment and palliative care are not mutually exclusive, especially in a
disease that is at the same time chronic and life-limiting. In an integrated model, both
active treatment and palliative care are offered concurrently, beginning at the time of
diagnosis of an incurable illness (see Figure 2) (Kristjanson, 2001). In this paradigm, use
of palliative care strategies begins at the time of diagnosis and increases as the patient
nears death. This model is especially relevant to the care of cancer patients who begin
with active treatment in hopes of a cure and end life in hopes of a comfortable death.

A modification of the concurrent model of palliative care is suggested for the

ESRD population (see Figure 3). End-stage renal disease patients differ from other

 

 

Active Treatment Palliative
Care

 

 

 

 

Diagnosis Death

Figure 1. Traditional model of palliative care

 

Active Treatment

Palliative Care

 

 

 

Diagnosis Death

Figure 2. Integrated model of palliative care

 

Active Treatment

Palliative Care

 

 

 

Diagnosis Death

Figure 3. Suggested model of palliative care for ESRD patients

 

chronically ill individuals in that they are forced to confront end-of-life (EOL) issues
from the time of diagnosis, facing certain death unless they receive active treatment in the
form of RT. Even with RRT, these patients live with the reality of a shortened lifespan.
Data provided by the USRDS (2002) indicate that expected remaining lifetimes for
dialysis patients are one-quarter to one-fifth those of the general population. Furthermore,
adjusted first year mortality rates are 21.5% for dialysis patients, with rates remaining
remarkably constant through the fifth year of dialysis (Reikes, 2000). End-stage renal
disease patients also experience a multitude of unpleasant symptoms as a result of both
the illness and its treatment. It is argued that the integrated model of palliative care
requires modification for this unique patient population. In the revised version of the
model, active treatment and palliative care share equally in the treatment plan from the
time of diagnosis to death.
Both versions of the concurrent model are supported by the World Health
Organization (WHO) and are consistent with its definition of palliative care. According
to the WHO (2001),
Palliative care is the active total care of patients whose disease is not responsive
to curative treatment. Control of pain, of other symptoms, and psychological,
social and spiritual problems, is paramount. The goal of palliative care is the
achievement of the best QOL for patients and their families. Many aspects of
palliative care are also applicable earlier in the course of the illness in
conjunction with active treatment.

This definition depicts palliative care as a service that can be offered along with active

therapy as well as after the failure of curative treatment (Kristjanson, 2001).

The concurrent model of palliative care has not yet been accepted by the
healthcare community. In the present healthcare system, most palliative care services are
offered through hospice and still tend to be an option confined to EOL (V achon, 2001).
In addition, reimbursement for hospice benefits is tied to criteria that limit the ability of
ESRD hemodialysis patients to take advantage of these services that specialize in
symptom relief. To make use of the Medicare Hospice Benefit, patients must have a 6
month prognosis and forgo aggressive treatment (U. S. Department of Health and Human
Services Health Care Financing, 2003). Although the Healthcare Financing
Administration requires that prognosis be based only on the physician’s best knowledge
of the disease course, it is difficult to anticipate length of survival in this population
(Barriers to hospice, 2000). In addition, few patients would decline aggressive treatment
(i.e., dialysis) in exchange for hospice care. As a result of these two criteria. ESRD
patients are rarely referred to hospice for palliative care.

Either modification of the current Medicare criteria for admission to hospice or
the creation of a separate benefit to cover ESRD patients is needed to assure access to
palliative care services currently available. Yet another alternative is the employment of
clinicians that specialize in palliative care in both nephrologists’ offices and dialysis
clinics. Programs to educate the clinical dialysis staff in palliative care strategies might
also be put in place. The downside is that each of these options requires a financial
commitment at a time when economic resources are limited.

However, these expenditures can be justified in light of the costs associated with
unrelieved symptoms. Poorly controlled symptoms often result in increases in both the

utilization of health care resources and the economic costs of care (Wilson & Cleary, 1995).

Unmanaged symptoms are also expensive personally and financially in that they
challenge an individual’s capacity to engage in normal life activities, including
employment. Furthermore, living with a chronically ill individual suffering from
unrelieved symptoms affects the entire family unit in the struggle to maintain financial
stability and an acceptable QOL.

Symptom relief is a primary goal of care as well as an indicator of the quality of
care provided hemodialysis patients. Health professionals, who play a major role in
helping patients manage their symptoms and achieve a satisfactory QOL, have a vested
interest in advocating for changes that promote their efforts to achieve these objectives.
To effectively campaign for change, however, healthcare providers must be armed with
information that documents the need for palliative care services in the hemodialysis
population. Data that provide this support are not currently available in the literature.

Research to explore and describe the symptom experience of hemodialysis
patients is the first step in validating the need for specialized symptom management
services. To accomplish this goal, the current study will examine several areas of the
symptom experience of hemodialysis patients including: 1) the factors that influence
symptoms, 2) the symptoms themselves, 3) strategies used by patients to relieve
symptoms, 4) the level of relief achieved, 5) the relationships among symptoms, level of
relief, and QOL, and 6) the impact of level of relief on the relationship between
symptoms and QOL. The data that result from this study will either refute or confirm the
assumption made in this discussion that hemodialysis patients live with unrelieved
symptoms that affect their QOL. Should this claim be validated, the next phase, to

determine whether palliative care leads to improved symptom management and QOL,

will be set in motion.
Theoretical Model
Theory of Unpleasant Symptoms

The theoretical model for this study is based on the middle range theory of
symptom experience, The Theory of Unpleasant Symptoms (TOUS) (see Figure 4).
Therefore, a summary of the TOUS will be presented first, followed by a description of
the modified model that was the foundation for this investigation. The TOUS
hypothesizes that symptom experience is comprised of influencing factors that cause or
affect the nature of the symptoms, the symptoms themselves, and their consequences
(Lenz, Pugh, Milligan, Gift, & Suppe, 1997). Further clarification of the model is
presented below, beginning with a discussion of factors thought to influence symptom
experience.

Influencing factors. The TOUS proposes that three categories of variables affect
the occurrence and perception of symptoms; physiological, psychological, and situational
(Lenz et al., 1997). Examples of physiological factors include electrolyte imbalance, the
presence of pathological conditions, and altered nutritional status. The psychological
category is comprised of variables such as mood, affect, and mental states of depression
and anxiety. Situational factors refer to aspects of the social and physical environment
that affect symptom experience. Examples include marital status, level of education,
social support, environmental air quality, heat, and humidity. According to the TOUS, all
three categories of influencing factors are interrelated and interact with one another to
shape the overall symptom experience (see Figure 4) (Lenz et al., 1997).

Symptoms. Symptoms are conceptualized as multidimensional phenomena defined

10

Influencing Factors

Physiologic H Psychologic

Factors Factors

\ /

Situational
Factors

 

Key

 

Influences

Interacts with

 

Reciprocal feedback -------

Figure 4. Theory of Unpleasant Symptoms

‘0.00‘°..

duration
distress
quality
Symptom 1
Symptom 2

distress intensity
quality

Symptom

intensity

Umé

. . . . o 0 Performance
0

I

 

 

Note. From Lenz, E. R., Pugh, L. C., Milligan, R. A., Gift, A., & Suppe, F. (1997). The

middle-range Theory of Unpleasant Symptoms: An update. Advances in Nursing Science,

19(3), 14—27. Copyright 1997 by Lippincott Williams & Wilkins. Reprinted with

permission.

11

by intensity, timing, distress, and quality (Lenz et al., 1997). Intensity refers to
severity of the symptom. The time dimension includes both the frequency of 0c
and the duration of the symptom. Distress is the degree to which a person is bot
symptom. Lastly, quality is the way in which a symptom is manifested and ofte
by the descriptors used to characterize it (Lenz et al., 1997).

Symptoms rarely occur in isolation. More often, patients experience mu.
symptoms simultaneously (Dodd, Miaskowski, & Paul, 2001b; Lenz et al., 199'.
TOUS asserts that the perception of a single symptom is different than the sense
the same symptom when it occurs in combination with others. Co-occurring syr.
are thought to interact to produce an experience that is multiplicative rather that.
(Lenz et al., 1997).

Performance. The final component of the TOUS is performance, the con
of the total symptom experience. Performance includes both functional and cogr
activities (Lenz et al., 1997). Functional performance is defined as physical fimc
social functioning, and the ability to accomplish normal role related tasks. Cogn
performance refers to the capacity to think, concentrate, and problem solve. The
suggests that the presence of symptoms results in reduced performance ability (1
al,1997)

Modified study model.

The model that guides this study is an adapted version of the TOUS. Rex
made in the TOUS resulted from both a review of the literature and a unique
conceptualization of the symptom experience that includes symptom relief (see I

Influencing factors. This component of the TOUS was included in the st

12

.328 use mesa: E can mEoEEIAm “Gamma—AED mo @025. 65. 3 mcofiomEo—Z ”362

$5an $336080: mo 85:35 8038.3 05 mo 3608 33382? .m oSmE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

\mamte iv than: $3. flmbflc so 0:5 mo cameo..— -
ousuuQ aeuoebu‘eh cosmos? mo $54 .
mesa .852 .
/ \ DUMM— I
coccoo .
ufluuuehaxwamz ow< .
ofiafiwofioggogém
:23. :3me no .3: .3
mMOHO<m 025753152“
unashaoaeb 585$? .
$055 EBA—Eta . a
REESE .. cormSQ seamen on .
oufieaeom .33 Bash . : chosen—ouch c033 .
BSEQmSeSMSefiAfi _. bagom Eomwo—onotnmm
ufizeauzak 33 «Eric . . 9:» .
Sis Seam coastsooo as m 955E oEoZmDEE
m SOFA—firm
mozgogmm

 

 

 

 

 

.8333 £9:me -
802980 29888
$938080: 023808

“5:33 0885 ..

32335563

 

mMOHO<m 02.52ng

 

 

 

l3

model unchanged. As in the TOUS, three categories of factors influence the presence

and perception of symptoms, including the physiological, psychological, and

situational/ demographic. Physiologic factors of interest include metabolic homeostasis
and numbers of comorbid conditions as indicators of disease severity. Dialysis adequacy,
a gauge of the effectiveness of hemodialysis treatment, is also included as a physiological
factor that potentially influences symptoms. Anxiety and depression represent
psychological influencing factors. Finally, age, gender, race, marital status, level of
education, and length of time on dialysis are the situational factors that are the focus of
this study.

Symptoms. The symptom component of the TOUS was modified following a
review of the literature. As depicted in the revised study model, the dimensions of
symptoms include: occurrence and distress (McDaniel & Rhodes, 1995; Rhodes &
Watson, 1987). Symptom occurrence includes the following dimensions: severity
(intensity), frequency (how often the event occurs within a given time frame), and
duration (length of time the event lasts) (McDaniel & Rhodes, 1995). Symptom distress
is the degree or amount of physical and/or mental upset experienced from a specific
symptom Glhodes & Watson, 1987). Both occurrence and distress describe the
experience of individual symptoms (Rhodes & Watson, 1987).

An additional dimension, concurrence, has been added to the model. Concurrence
describes the coexistence and interaction of symptoms and, together with occurrence and
distress, completes the multidimensional conceptualization of symptoms. Concurrence is
considered an important aspect of symptoms but it is not labeled as a separate dimension

in the TOUS (Lenz et al., 1997). Doing so emphasizes the importance of the interaction

14

 

of multiple symptoms in shaping the overall symptom experience.

Level of Symptom Relief

Another modification of the TOUS is the inclusion of level of symptom relief as a
significant component of symptom experience in ESRD. Symptom relief strategies,
degree of relief achieved, and perceived satisfaction with relief define this component of
the model. Symptom relief strategies are the methods used by patients to achieve a level
of symptom relief. Strategies include those that are prescribed by healthcare professionals
as well as any that are self initiated. Degree of relief is defined as a subjective evaluation
of the extent to which symptoms are relieved. Perceived satisfaction with relief is the
extent to which the individual is content with his/her current level of relief from
symptoms. In the study model, level of symptom relief moderates the effect of symptoms
on QOL.

Performance. The TOUS includes the ability to perform functional and cognitive
activities as the consequence of symptom experience. However, in the current study, this
component has been amended to include a more global performance measure, QOL.

F errans’( 1997) conceptualization of QOL is integrated into the study model as the
chosen performance indicator. Based on the View that QOL is a subjective concept
dependent on the individual’s perspective, Ferrans (1997) defined QOL as “a person’s
sense of well-being that stems from satisfaction or dissatisfaction with the areas of life
that are important to him/her” (p. 113).

Ferrans’ model encompasses four domains: health and functioning,
psychological/spiritual, social and economic, and family. The health and functioning

domain includes the following aspects of life: usefulness to others, physical

15

independence, ability to meet family responsibilities, own health, energy, worries, contrc
over own life, leisure time activities, potential for a long life, ability to travel, sex life,
health care, and potential for a happy old age. Satisfaction with life, happiness in general
satisfaction with self, achievement of personal goals, peace of mind, personal appearance
and faith in God are components of the psychological/spiritual domain. Elements of the
social and economic domain are standard of living, financial independence, home,
neighborhood, job/unemployment, friends, emotional support from others, and education
Finally, family happiness, children, relationship with spouse, and family health are
factors that comprise the family domain. QOL is a product of both the degree of
satisfaction with aspects of life in each domain as well as the perceived importance of
those aspects of life to the individual. The overall study model predicts that symptoms
directly affect QOL by their influence on an individual’s satisfaction with areas of life
that he/she considers important.

In summary, the TOUS and the relationships it proposes among influencing
factors, symptoms, and performance provide the foundation upon which the present stud;
model was built. The theory hypothesizes that a variety of physiologic, psychologic, and
situational/demographic factors potentially influence symptoms which, in turn, affect
performance. For the current study, however, the TOUS was modified to include level of
symptom relief as a significant component of total symptom experience. The dimension
of concurrence was also specifically labeled to emphasize its importance to the overall
symptom experience. The final model is presented in Figure 5.

Symptom Experience of Hemodialysis Patients

The current state-of-knowledge regarding the symptom experience of

16

hemodialysis patients will be discussed within the study’s theoretical framework
described in the previous section. An overview of studies related to each component of
the model will be presented. In some instances research is limited, reinforcing the
significance of the present investigation.
Influencing Factors

A variety of physiological, psychological, and situational/demographic factors are
thought to precipitate and/or modify the symptoms hemodialysis patients experience.
A select few are of particular interest to this study including: disease severity and dialysis
adequacy (physiological); anxiety and depression (psychological); and age, gender, race,
marital status, level of education, and length of time on dialysis (situational/demographic).
The relationship between some of these variables and symptoms has not been previously
explored in the ESRD population. Others have been examined but with conflicting results.
Influencing Factors: Physiological

As depicted in the study model, it is hypothesized that both disease severity and
dialysis adequacy affect symptoms. The term ‘disease severity’ is defined as the degree
of physiological disturbance present (Craven, Littlefield, Rodin, & Murray, 1991). The
degree of metabolic homeostasis and the presence of comorbid conditions are indicative
of overall physiological disturbance (Craven et al., 1991). Both are included as
determinants of disease severity in the current study.

Adequacy of dialysis refers to the delivered ‘dose’ of dialysis, namely the amount
0f effective dialysis a patient achieves in one treatment session (Curtis, 2004). The
measurement of urea clearance, as expressed by Kt/V (i.e., dialyzer clearance x time /

Volume of body water), has been widely accepted as one marker of dialysis adequacy

l7

 

(National Kidney Foundation [NKF], 2001). Kt/V represents the rate at which urea, a
waste product, has been removed from the blood. Consistent with national guidelines,
Kt/V is the indicator of dialysis adequacy chosen for this study.

Metabolic homeostasis. Metabolic homeostasis refers to the body’s ability to
maintain relative constancy of the internal environment (Anderson , Anderson, & Glanze,
1998). Laboratory values are measures of the body’s ability to maintain homeostatic
balance and one gauge of disease severity (Griffin, Friend, & Wadhwa, 1995). Some
research has examined the effect of biochemical imbalances on the symptoms
experienced by hemodialysis patients. However, findings are inconclusive,
particularly in relation to the commonly occurring symptoms of fatigue and problems

sleeping (Brunier & Graydon, 1992; Cardenas & Kutner, 1982; McCann & Boore, 2000).
Although fatigue is one of the most prevalent symptoms reported by hemodialysis
patients, its causes are not well understood (Srivastava, 1989). The toxic effects of uremia
combined with low hemoglobin levels that are typical of ESRD are often implicated as
precipitating factors. However, research has not consistently demonstrated the
relationship between these laboratory parameters and fatigue. McCann and Boore (2000)
found no association between biochemical variables, including urea and anemia, and
fatigue in their sample of 50 hemodialysis patients. Likewise, anemia did not contribute
Significantly to the fatigue reported by 43 patients in another investigation conducted by
Brunier and Graydon (1992). These findings corroborated those of an earlier study by
Cardenas and Kutner (1982), who observed no correlation between fatigue ratings and
hematocrit or any of the other lab values examined, including BUN and creatinine.

Some research related to sleep disturbances also contradicts the conventional

18

thinking which links uremic toxins to disordered sleep patterns (Daugirdas & Ing,
The sleep patterns of 54 hemodialysis patients were examined by Walker, Fine, ant
Kryger (1995) and no correlation was found between urea and creatinine levels anc
problems sleeping.

In a descriptive study of 191 ESRD patients, Barrett and colleagues (1990)
note a significant correlation between sleep disturbances and urea levels. However
strength of the association (r = 0.19, p3 0.05) called into question the clinical rele‘
of this finding. Additional significant correlations between laboratory indices and
symptoms were demonstrated in this study as well. Levels of alkaline phosphatase,
calcium, and uric acid were positively related to joint pain. Serum bicarbonate leve
negatively correlated with cramps. As was the case with urea and sleep disturbance
strength of the relationship among these variables was low.

Finally, Wolcott, Nissenson, and Landsverk (1988) examined the relationsl
between laboratory values and symptoms using global measures of each of the vari
The Chemistry Abnormality Score (CAS) summarizes the results of 13 routinely
performed biochemical tests in dialysis patients. The Active Clinical Problems Scc
(ACPS) summarizes 36 symptoms/signs and recent hospitalizations. No significan
correlation was noted between these two scores.

Based on the research available, it is not yet clear whether there is a signifii
relationship between biochemical indices and the symptoms reported by hemodialj
patients. It seems reasonable to assume that disturbances in homeostatic balance m

Precipitate somatic symptoms. However, this proposed connection requires further

investigation.

19

Comorbid conditions. The presence of comorbid conditions is the second
determinant of disease severity of interest to this study. Comorbidity is defined as a
significant concurrent disease in addition to ESRD which involves organs other than the
kidneys but which may also be responsible for the renal failure (Khan, 1998). Again,
research linking comorbidity and symptoms is sparse and inconsistent.

Merkus and colleagues (1999) examined the relationship between a variety of
clinical characteristics and physical symptoms in a sample of 120 chronic hemodialysis
patients. The clinical variables included the underlying kidney disease, lean body mass, a
selection of laboratory values, hydration status, residual renal function, and comorbidity.
Of these factors, only a medium to high comorbidity-age index was significantly
associated with a higher symptom scores. This index, based on age and presence of
comorbid conditions, indicated that patients over the age of 70 years with one or more
comorbidities experienced greater symptom burden.

Conversely, comorbidity was not among the significant correlates of somatic
symptoms reported by Barrett, Vavasour, Major, & Parfrey (1990) in an earlier study of
73 hemodialysis patients. There was also no correlation found between comorbid
conditions, such as cardiovascular disease, and fatigue in a study conducted by Cardenas
and Kutner (1982).

Dialysis adequacy. Dialysis adequacy is also thought to play an influential role in
shaping the symptoms that hemodialysis patients experience. When kidney failure occurs,
the body loses its ability to remove the toxic end-products of metabolism. As waste

products accumulate, symptoms of the uremic syndrome appear. Hemodialysis is

instituted to compensate for loss of renal function, restore homeostatic balance, and

20

 

reduce unpleasant symptoms. The effectiveness with which hemodialysis performs
functions is quantified by Kt/V, as defined earlier in this discussion. The current gt
standard is a Kt/V _>_ 1.2 (NKF, 2001). The logical deduction is that Kt/V values be
that norm would be associated with increasing symptoms of uremia. Although rese
does not consistently validate this hypothesis, the relationship has been demonstrat
Morton and colleagues (1996) examined the relationship between dialysis
adequacy and QOL in a sample of 55 hemodialysis patients. Measured using the
subscales of the Rand 36 Item Health Survey and Kt/V values, the correlations bet
fatigue/energy and pain subscales and Kt/V were not significant. Likewise, in the s
conducted by McCann and Boore (2000) and Merkus et a1. (1999) previously discr
dialysis adequacy was included as one of the biochemical variables examined. Nei'
investigation found a significant relationship between Kt/V and symptoms.
Data presented by Hamilton and Locking-Cusolito (1998) contradict these :
In a pilot study that followed 11 hemodialysis patients for 3 months following initi
of treatment, symptom burden decreased with adjustments in therapy that producer
increased Kt/V values. Degree of fatigue also declined. The authors cautioned that
factors not examined may have also contributed to improved symptom status. In ac
the small sample size limited their ability to state conclusively that more effective
dialysis was associated with lower symptom burden.
However, studies examining the benefits of daily hemodialysis provide add
evidence of the relationship between dialysis adequacy and symptoms. Recent rese
has demonstrated that Kt/V levels increase with short daily treatments. Improveme

intradialytic symptomatology (i.e., nausea, cramps, and dizziness), post-dialysis fat

21

and uremic symptoms have accompanied the rise in Kt/V values (Curtis, 2004; Lindsay,
Kortas, & the Daily/Nocturnal Dialysis Study Group, 2001; Maduell et al., 2003). Since
none of these studies included sample sizes greater than 10, careful interpretation of these
findings is warranted. Regardless, these studies justify inclusion of dialysis adequacy as
an influencing factor in this study.

The research presented illustrates the inconsistencies in the literature surrounding
the relationship between physiologic factors (i.e., disease severity and dialysis adequacy)
and symptoms experienced by hemodialysis patients. Neither disease severity nor dialysis
adequacy has been established unequivocally as factors that do or do not influence
symptoms. Therefore, these variables were explored in this investigation in an effort to
further clarify their connection.

Influencing Factors: Psychological

The second category of factors that are hypothesized by the study model to
influence symptoms is the psychological. The interaction of an individual’s state of mind
and physical condition has long been established in the literature. Symptoms are defined
as subjective indicators of change in a condition as perceived by the individual
(Anderson, Anderson, & Glanze, 1998). Psychological status is the filter through which
these subjective perceptions are interpreted. Patient reports of symptoms are outcomes of
these perceptions (The University of California, San Francisco School of Nursing
Symptom Management Faculty Group, 1994; Dodd et al., 2001a).

Anxiety and depression are two disorders that influence the perception and
interpretation of symptoms. Previous research has verified the relationship between these

mental states and the symptoms experienced by persons living with chronic illnesses

22

 

(Simon, 2001). Since anxiety and depression are commonly noted in the ESRD
population, their inclusion as psychological influencing factors is critical to
understanding the symptom experience of hemodialysis patients (Christensen & Ehll
2002; Kimmel, 2002).

Anxiety and depression have been linked to specific symptoms such as fatigt
Significant correlations between both disorders and fatigue were noted by McCann 2
Boore (2000) in a study of 25 hemodialysis patients. To eliminate the confounding 0
physical symptoms of ESRD and mental state, anxiety and depression were assessed
using the Hospital Anxiety and Depression Scale (HADS). This tool includes items I
solely on psychic symptoms of neurosis and excludes the somatic symptoms that co'c
with physical illness (Zigmond & Snaith, 1983). The correlation between anxiety an:
depression and fatigue was demonstrated despite the existence of physical illness in
study group (McCann & Boore, 2000).

Cardenas and Kutner (1982) measured depression but not anxiety in their stu
fatigue in 137 dialysis patients. They concluded that depression is a principal cause (
fatigue. This inference was supported by the fact that patients who experienced
significant fatigue upon arising, when its level would be expected to be its lowest, a1
had scores that indicated significant clinical depression.

Depression and anxiety have also been found to be associated with overall
SB’mptom experience. Killingworth and Van Den Akker (1996) examined the relatior
between mental state and symptom status in a sample of 170 dialysis patients. Using
HADS to measure anxiety and depression, they found significant correlations betwei

both the anxiety and depression subscales and uremic symptoms that accompany ren

23

failure. As noted previously, the HADS does not include somatic symptoms commonly
experienced by physically ill individuals. The assessment of anxiety and depression in
this sample was based solely on the presence of psychological symptoms.

Likewise, in their study of 73 hemodialysis patients, Barrett et al. (1990)
examined the connection between psychological factors and the severity of eight somatic
symptoms including: tiredness, sleep disturbance, cramps, pruritis, headache, nausea,
dyspnea, and joint pain. Affect was measured using a scale comprised of 11 emotions that
the researchers considered of importance in ESRD, among them feeling sad and scared.
The severity of each somatic symptom Was significantly related to indices of affect.
Although the affect scale used was not a specific measure of anxiety or depression, it was
able to demonstrate a relationship between psychological well-being and symptoms
patients experienced.

The empirical evidence corroborates the presence of a relationship between
psychological status and symptoms in the ESRD population. Anxiety and depression
have been consistently linked to symptom status. Whether anxiety and depression
intensify the perception of symptoms or anxiety and depression are the result of symptom
burden remains uncertain. However, there exists a clear link that validates the inclusion
0f anxiety and depression in the theoretical model for this study.

Situational/demographic factors. Examination of the connection between
situational/demographic variables and symptoms has not been a major focus of previous
research. However, a number of these variables conceivably affect reporting of

SYmptoms. Of interest to this study are age, gender, race, marital status, level of

education, and length of time on dialysis.

24

Age is frequently reported to describe the study sample and not often considered a
major variable of interest. Research investigating the effect of age on the symptoms
reported by hemodialysis patients is sparse. Age was positively correlated with symptom
burden when measured in combination with comorbidity in the study by Merkus et al.
(1999) mentioned earlier. Older age and greater comorbidity were associated with higher
levels of symptomatology. The effect of age alone was not reported. Similarly, Barrett et
al. (1990) found that age was related to reports of headache and nausea in their study of
the correlates of somatic symptoms in patients on dialysis. Age was not a major focus of
either of these studies.

Gender. Like age, the impact of gender on the symptoms experienced by
hemodialysis patients has not been documented empirically. Only one study was found
that included gender in the analysis of symptom reporting. In their study of fatigue in
dialysis patients, Cardenas and Kutner (1982) found that average fatigue ratings were
similar for men and women. No other studies specific to ESRD were located that
examined the relationship between gender and symptoms. It is yet to be determined if
symptom response is gender specific in the hemodialysis population.

Race. Investigation of the interaction of race and symptoms has been similarly
neglected. In the only study found that examined the differences in the symptoms
reported by race, Kutner and colleagues (2000) discovered that Blacks were significantly
less likely than Whites to complain of nausea, sexual dysfunction, and general fatigue.
Black patients also reported less post dialysis fatigue and fewer hours needed for
recovery following treatment. No other racial groups were included in this study.

Research that investigates the relationship between race and symptoms is notably lacking

25

in the nephrology literature.

Marital status and level of education. Research relating marital status and level of
education with symptoms is sparse, with these variables most often reported only as
descriptors of sample demographics. However, Merkus et al. (1999) did include these
variables in their analysis but did not find an independent association between them and
the symptoms reported by hemodialysis patients. McCann and Boore (2000) also failed to
observe any relationship between marital status of dialysis patients and level of fatigue.
Sample size in both studies was comparatively small, which may account for the lack of
significant findings. Because marital status and level of education have not been
adequately studied, their influence on symptoms warrants further investigation.

Length of time on dialysis. It has been suggested that the experience of ESRD can
be described in relation to an illness trajectory, with different phases emerging over time
(J ablonski, 2004). Each phase is thought to be characterized by unique physical and
psychological events, tasks, and shifting demands (Corbin & Strauss, 1992; Rolland, 1987).
In light of this conceptualization, the experience of dialysis, as well as the symptoms that
accompany it, might be expected to vary over time.

Neither the concept of an ESRD illness trajectory nor the effect of time on
dialysis on symptoms has been researched. Only two studies were found that examined
the correlation between length of time since initiation of treatment and symptoms and in
neither was it a major focus. In their sample of 137 hemodialysis patients, Cardenas and
Kutner (1982) reported that those on hemodialysis for greater than 4 years experienced
lower levels of fatigue than those treated for a shorter time. However, in a smaller sample

of 39, McCann and Boore (2000) did not observe this relationship. Additional research is

26

needed to establish whether or not a relationship exists between time since initiation of
dialysis and symptoms.

The lack of studies examining the influence of situational/demographic variables
on self-reported symptoms is striking. Furthermore, this discussion of all of the categories
of influencing factors, especially the physiological and situational/demographic, has
illustrated the need for clarification of their connection to symptoms. Although many of
the factors included in the present study are not modifiable, knowledge of those factors
that are related to higher levels of symptoms may help detect patients at risk for higher
levels of symptoms. Identification of those factors that are modifiable may help to
pinpoint areas for intervention as well.

Symptoms

Research indicates that ESRD patients suffer from an array of troublesome
symptoms. Some of them are triggered by specific organic diseases, such as dyspnea due
to chronic lung disease or joint pain caused by gout. Others are difficult to attribute to a
specific organic pathology (e.g., problems sleeping, fatigue, and headache). Still others
are related to treatment side effects (e. g., cramps and nausea).

Regardless of etiology, a number of studies have identified the symptoms that are
commonly experienced by hemodialysis patients. Fatigue is often the most prevalent

symptom reported (Barrett et al., 1990; McCann & Boore, 2000; Merkus et al., 1999;
Parfrey et al., 1987; Parfrey, Vavasour, Henry, Bullock, & Gault, 1988; Parfrey et al.,
1989). Other symptoms that have been found to occur in greater than 20% of the ESRD
population include: itching, headache, sleep disturbance, cramps, pain (joint, chest,

abdominal), shortness of breath, nausea/vomiting, and muscle weakness (Parfrey et al.,

27

 

1988)

Although research has documented the specific symptoms that are most common
among dialysis patients, an exploration of the multidimensional nature of symptoms is
yet to be completed. Studies have focused primarily on the presence/absence of
symptoms and their associated severity or frequency of occurrence. Other dimensions of
symptom experience, such as distress and concurrence have been overlooked or
inconsistently described.

McCann and Boore (2000) measured only the severity of symptoms in their study
of fatigue. Hemodialysis patients reported tiredness as the most severe symptom. Muscle
weakness, joint pain, and lack of sleep rounded out the top four symptoms in terms of
severity. Likewise, Parfrey and colleagues (1987) examined the severity of symptoms
experienced by 107 hemodialysis patients. Tiredness was once again rated the most
severe, followed by cramps, headaches, itching, and sleep disturbances in descending
order.

Killingworth and Van Den Akker (1996) also described only one symptom
dimension, in this instance, frequency of occurrence. Greater than 20% of the
hemodialysis patients in their study experienced tiredness, itching, sleeping problems,

and muscle weakness on most days. Apparent from this and studies of symptom severity,
tiredness is one of the most problematic symptoms for these patients.

Multiple dimensions of symptoms were intentionally assessed in only one study
found in the literature. Parfrey et al. (1988) expanded a physical symptom scale
developed in an earlier study to incorporate a number of clinical features of symptoms,

among them severity, duration, and frequency of occurrence. For example, it was noted

28

 

that fatigue occurred every day in 65% of the sample and lasted for greater than 6 hours
in 69% of those patients. Itching was reported to be a daily event for 70% of the patients,
with a reported duration of over an hour for most of them (80%). These expanded
descriptions of symptoms generate a more thorough understanding of the total
experience. Distress and concurrence were not examined in this study.

A relatively new addition to the conceptualization of symptoms, concurrence has
not been explored in studies of dialysis patients. However, concurrence was alluded to by
McCann and Boore (2000) in their study of fatigue. On average, hemodialysis patients in
their sample reported seven symptoms. There was a significant positive correlation found
between fatigue severity scores and the number and severity of additional symptoms.
This observation provides some evidence of an interaction among concurrent symptoms
However, whether the nature of the interaction is multiplicative or additive is not evident
from these findings.

Although the specific symptoms hemodialysis patients experience are well
documented, their multidimensional characteristics are not. Previous investigations have
focused on limited aspects of symptoms, typically occurrence and severity or frequency.
Given the recent introduction of concurrence as a symptom dimension, it is not surprising
that no research was found describing this dimension in ESRD. Consequently, a complete
picture of symptoms is not yet available.

Level of Symptom Relief
As depicted in the theoretical model for this study, symptom relief strategies,
degree of relief, and satisfaction with relief interact to define overall level of symptom

relief. It is proposed that level of relief then moderates the impact of symptoms on QOL.

29

 

However, this relationship has not been documented in the literature. Research
investigating the extent to which ESRD patients obtain relief from their symptoms is
virtually nonexistent. Typically, the focus is on the presence/absence of symptoms rather
than the amount of relief obtained or the methods used to manage them (Hoffart, 1995).

Only one investigation was found in which patients were questioned about
symptom relief. In a study discussed previously in relation to symptom dimensions,
Parfrey et al. (1988) asked dialysis patients several questions about each symptom they
reported. For some of the symptoms, patients were asked if they took medications for
relief. However, neither the adequacy of the relief nor the effectiveness of other
interventions was investigated. Failure to investigate symptom relief results in a
significant gap in knowledge of the total symptom experience associated with ESRD.
Performance

QOL is the final component of the symptom experience model that is the
theoretical foundation of this study. QOL is a global performance indicator that
represents overall satisfaction with the physical, psychological/spiritual, social/economic,
and family aspects of life as well as their relative importance. The model depicts a direct
link between symptoms and QOL, suggesting that symptoms diminish QOL by impacting
areas of life that are important to an individual. Research has validated the positive
relationship between symptoms and QOL proposed by the model.

Killingworth and Van Den Akker (1996) assessed the QOL of 70 hemodialysis
patients using the Quality of Life Index—Dialysis Version (QLI) (Ferrans & Powers,
1985). Patients’ total QLI scores were skewed toward the higher end, suggesting

satisfactory QOL. They noted, however, a significant positive correlation between

30

symptom severity and QOL. Those patients with the most intense symptoms reported the
lowest level of QOL. Because the correlations between symptom scores and the subscales
of the QLI were not reported, the influence of symptoms on different aspects of life
cannot be determined.

The influence of symptoms on QOL was also assessed by Merkus et al. (1999) in
a study of 120 incident chronic hemodialysis patients. Symptom burden was shown to be
an independent determinant of diminished QOL, measured using the MOS SF-36.
Physical component summary scores (PCS) and mental component summary scores
(MCS) were calculated to summarize the eight scale profile. Symptom burden explained
17% of the variance in the PCS scores and 23% of the variance in MCS scores.

Srivastava (1989) examined the characteristics of chronic fatigue experienced by
27 hemodialysis patients and the impact on several aspects of functioning. A significant
relationship between fatigue and QOL was found, with a higher levels of fatigue linked to
lower perceived QOL. The majority of patients who reported a QOL of less than 90
(possible scores ranged from 0-100, low to high quality) also indicated that their QOL
would improve appreciably if their tiredness was relieved.

The effect of symptoms on QOL was explored in a similar manner by Parfrey
et al. (1988). In their study of the clinical features of six somatic symptoms (i.e.,
tiredness, cramps, itching, shortness of breath, headaches, and joint pain), patients were
asked whether relief of each of the symptoms would improve their QOL byZ 10%.
Patients indicated that relief of tiredness, cramps, and sleep disturbances would
significantly improve their feelings of well-being.

Although a variety of factors undoubtedly interact to shape QOL, empirical

31

evidence hints at the important influence of symptoms. In studies that have included
symptoms as a correlate, the negative relationship between symptoms and QOL has been
consistently demonstrated. The conclusion that symptoms diminish the perceived
well-being of dialysis patients seems warranted, validating that link in the study’s
theoretical model.
Summary

Apparent from this review of the nephrology literature, the total symptom
experience of hemodialysis patients has not been fully described. The present study
contributes to the knowledge base by documenting the relationship that exists between
specific physiological (disease severity and dialysis adequacy), psychological (anxiety
and depression), and situational/demographic (age, gender, race, marital status, level of
education, and length of time on dialysis) factors and the experience of symptoms. The
literature abounds with inconsistencies regarding the most significant of these factors.

This research is also the first to examine the multidimensional nature of
symptoms reported by hemodialysis patients. The level of relief that patients obtain from
their symptoms will also be explored, adding unique information to the current
understanding of symptom experience in ESRD. Finally, the ability of hemodialysis
patients to maintain a satisfactory QOL is a major concern of healthcare professionals.
Factors that are associated with diminished QOL will be identified to help direct their
efforts to improve the lives of hemodialysis patients, especially in terms of improved
symptom relief.

Ultimately, this study responds to an appeal made by the End-Stage Renal

Disease Workgroup, a subgroup of The Robert Wood Johnson Foundation program

32

 

Promoting Excellence in End-of—Life Care (Moss & the End-Stage Renal Disease
Workgroup, 2001). After extensive study of the current state-of-care, this group called for
increased efforts to improve the lives of ESRD patients through the integration of
palliative care within dialysis practice. The nephrology research community was
challenged to help accomplish this goal through empirical validation of the benefits of
palliative care to ESRD patients. Accepting this challenge, this investigation seeks to
document the symptom experience of ESRD hemodialysis patients as the initial step in
determining if there is a need for improved symptom relief. In doing so, it will justify

further exploration of the benefits of palliative care to this vulnerable group of patients.

33

Chapter 3
METHODOLOGY
A primary aim of this study was to document the total symptom experience of

ESRD hemodialysis patients including: the factors that influence symptoms, the
multidimensional characteristics of symptoms, and the impact of both on QOL. Unique to
this research, the level of relief patients obtained from their unpleasant symptoms was a
major focus as well. This information was deemed essential to determine if there was a
need for improved symptom relief and palliative care in this patient population.

Research Questions

The following research questions were asked to achieve the research objectives:

1. What physiological, psychological, and situational/demographic influencing
factors predict the level of symptoms experienced by ESRD hemodialysis patients?

2. What are the characteristics of symptoms reported by ESRD hemodialysis
patients in relation to the following symptom dimensions: symptom occurrence (i.e.,
severity, frequency, duration), symptom distress, and symptom concurrence?

3. What strategies are used by ESRD hemodialysis patients to relieve symptoms?

4. What level of symptom relief is achieved by ESRD hemodialysis patients?

5. What is the relationship between influencing factors and symptoms and the

QOL reported by ESRD hemodialysis patients?

6. What is the moderating effect of level of symptom relief on the relationship

between symptoms and QOL?

A cross-sectional study design was chosen to answer the research questions. This

design permitted the inclusion of a wide variety of patients of varying ages, dialysis

34

histories, and disease severity, resulting in a description of the symptom experience of a
diverse group of patients. It was the intent of this study to provide a real-time snapshot of
symptom experience, not to document changes over time.

Sample

Patients receiving treatment in two in-center hemodialysis clinics located in
Grand Rapids, Michigan were recruited for this study. Both centers are privately owned.
At the time of the study approximately 174 patients were being treated at Clinic A. Of
those, 77% were White, 20% were Black, and 3% included other groups (e. g., Hispanic,
American Indian) (M. Larson, personal communication, June 3, 2003). Clinic B census
included 107 patients. Racial mix was nearly 50% Black and 50% White. Only a very
small portion of the patients were of other racial origins (R. Ploch, personal
communication, August 15, 2003). Both clinics treated patients on a Monday-
Wednesday-Friday or Tuesday-Thursday-Saturday schedule. Only Clinic A offered
evening dialysis sessions.

Participants recruited for the study had to be at least 21 years of age, alert and
oriented, able to understand and speak English, and competent to give informed consent.
Ability to read English was not required since measurement instruments were
administered via interview. In addition, patients had to be hemodynamically stable at the
start of the treatment and remain so throughout the interview. Otherwise, the interview
was terminated and completed during the next dialysis session when stable. Nurses
supervising the dialysis treatments monitored vital signs at least every hour and more
often as necessary.

Children under the age of 21 years were excluded from the study. ESRD is not

35

common in children, less than 0.4% of all dialysis patients in Michigan are below the ag
of 20 years (Renal Network of the Upper Midwest, Inc., 2001). In addition, children Wht
require renal replacement therapy (RRT) are typically treated with peritoneal dialysis or
transplantation.

The total convenience sample consisted of 130 patients, ranging in age from 22 t
88 years. Males (51%) and females (49%) were equally represented. The majority of
patients were White (69%). Black patients (27%) comprised the largest proportion of
those remaining, with only 4% Hispanic and American Indian patients included in the
sample. A slight majority of patients were married (52%). Sixty percent attended school
for 12 years or less and most were unemployed either because of disability (55%) or
retirement (28%). Mean length of time on dialysis was 32.25 months (range 042-240
months).

Demographic characteristics of each dialysis clinic are presented in Table 1. The
samples differed primarily in regard to race, reflecting their respective clinics. The racia‘
composition of the sample from Clinic A included 84% White, 12% Black, and 4%
Hispanic and American Indian participants. Fifty-three percent of the patients from Clin
B were Black, 43% were White, and 4% comprised other groups. No significant
differences were noted between the samples relative to other demographics including:
age, gender, marital status, level of education, employment status, and length of time on
dialysis. Forty-eight percent of all patients treated in Clinic A and 44% of the patients

dialyzed at Clinic B participated in the study.

The demographics of the sample, with a nearly equal percent of males and

females, reflected the in-center hemodialysis population in Michigan with respect to

36

Table 1

Demographic Characteristics of Patients by Clinic and Total Sample

 

 

Clinic A (n = 83) Clinic B (n == 47) Tot:
Characteristic n (%) n (%) n ("/t

Gender

Female 42 (51) 21 (45) 63 (4

Male 41 (49) 26 (55) 67 (5
Race

White 70 (84) 20 (43) 90 (6!

Black 10 (12) 25 (53) 35 (2'

Hispanic/American Indian 3 (4) 2 (4) 5 (4)
Marital status

Married 47 (57) 21 (45) 68 (52

Single, never married 6 (7) 9 (19) 15 (12

Divorced/separated 18 (22) 13 (28) 31 (24

Widowed 12 (14) 4 (8) 16 (12
Employment status

Employed fiill/part time 11 (13) 2 (4) 13 (1(

Retired 26 (31) 10 (21) 36 (25

Disabled 43 (52) 29 (62) 72 (Sf

Not employed/reason 3 (4) 6 (13) 9 (7)

unrelated to health

 

37

Table 1 (continued)

 

 

 

Clinic A (n == 83) Clinic B (n = 47) T 3
Characteristic n (%) n (%) nu
Level of education
Less than high school 22 (27) 14 (30) 27
High school diploma 26 (31) 16 (34) 42
Some college/vocational 28 (34) 16 (34) 44
school
College degree 6 (7) 1 (2) 7
Professional/graduate 1 (1) 0 1
degree
Length oftime on dialysis 27.88 (29.37) 39.96 (42.23) 32.25
in months
M (SD)

 

38

gender. Racial mix in Michigan is also evenly distributed; 50% White, 48% Black (Renal
Network of the Upper Midwest, Inc., 2001). It was necessary to recruit participants from
both clinics to achieve a more representative racial balance.

Because of the Health Insurance Portability and Accountability Act of 1996
(HIPPA), characteristics of those patients who did not participate in the study were not
available for comparison. Regulations enacted in April 2003 prior to the beginning of
data collection prevented access to private information regarding patients who did not
take part in the study (U. S. Department of Health and Human Services [USDHHS],
2003)

Measures

Structured interview questionnaires and a chart review form for laboratory tests
were used to collect data relevant to each research question. See Table 2 for a summary
of all instruments, including a brief description, scoring, and psychometric properties.
Copies of all measures appear in the Appendices.

Influencing Factors: Physiological

Disease severity. The term ‘disease severity’ was defined as the degree of
physiological disturbance present (Craven, Littlefield, Rodin, & Murray, 1991).
Determinants of overall physiological disturbance included degree of metabolic I
homeostasis and the presence of comorbid conditions (Craven et al., 1991). Both were
included as indicators of disease severity in the current study.

Metabolic homeostasis was defined as the ability of the body to maintain

constancy in the internal environment (Anderson et al., 1998). Of particular interest was

the kidney’s ability to maintain balance with respect to electrolytes, waste products, and

39

 

 

and n commoaom

.33 u 325
as n 28m 55
A933

9:03:88 5:066:00
sees 5:523
82550

58:00 56:3»

sowmgoaoom
beaconfi 96% com

82 seam as. seed
swam ease SQ

ASAUV 8< “58965::
©8833 32:20

o6. USN mufiwsofimm
succeeds mo owozou
05 c3 3:888 baa
Dogmaonmq ”325%va
“5:80 5825/

3: n 8338a
oTw H e>umomwzm
K6 H 38.8:
flmé owned 638
To co e88 mac:
R 63033 .852qu .
Hmé emcee ”2QO
To go v.83 was:
\1 6:833 30?ch .

cam 32:0 sad? \
08: x confide—o 35$me

concave 32:28
@5888 EN mo Sam

mmé H emcee 28m ._
N H BEBE?
a n 388% as?
o H Egon
“Sofie =< .
H u acetone“
0 H Echo:
83::ch .ZDm FM ..

 

moEoEonowmm

mcfioom

commmoaofl
838 2: Set 32:05 Be 323
magmas? oueaom .85qu _momwflonozwm
E bfioficfia E oommmoaoc 28m 8683ch ”E88.“
was 3258 Son wommomm< 28 No.23 358$ wfioqosufi
Sundown
>>M 38 3382305 minim
32:28 88583856 $82280
@5888 3 mo 65 ofimfiwofiom £80800
£83880:
£3802
c2588 _mommofimmcfim
85ch $8“ Eugenoofi Boom ”£88m
2 mo £38 2: :o comwm bzmgocfix DEEMED wfiocosmfi
eoumtomom 2332 03mg;

 

“assassins: hogs nee SERRA 85% no.9:

N 2an

40

 

3.0
u 808 .328 Bob
888-83 ”bagged

28:8 562a>

mud

H .808 SEEP? 39
Inc

M 88998? mo um :33
888-88 ”Em—Easom

€828 8:213

<2

fine So:
:ocofiwumm no mafia
83838 souoflmgm -
Eco 880:
«028 :m no mwcufi
mo E8 63038 mused .
was: EN :0 mwcufi mo
88 ”808 «0:8 30% .
288 2:25
$8: @808 So: comm .

omé u carom“:
o n @3898 Ho:
owcfi 88m .
208586
:08 Sm 808
wficwm an 8085:?
some 8m 38328
808 8899? 30B ..
2&8 EBaé a $8:
@808 58586 scam

<2

.8886 £2886

Susana SEQ/8 ”wfiufiofi
38586 Eowmém :80 HEB
38688 $28 was»: £3»
5503838 £226 c2638
«0:8 mo #96— ofi MGamm

:23 88688 8608558
_mofimxnm 2 26:8 8 8:2qu
>9 c8: 8&888 8888me

888%
can £03mb $80538.”—
5flo>8 8:622: amm
HEB 88688 8:83:58
32933 : mommomm<

88me so 08: mo 5mm»:
£03838 mo $32 .8838
358 603 .Hocaow .0?

”@68on ”Hosanna”: 38:00

 

moBoaonommm

matoom

coumuomom

mom—8 8995?
23m mozom Eoumgm mo #255
8885
>269 wad mEoHQEm mEoEEzm
oEmeoEoc
E32835
3858826 ӣ88m
oEmfiwofioQ macaosafi
8882

028$
/

abozmficoov N DENH

Ill

41

 

So u $8.8

mud

M 28288 28 Boom
36

H _mBEaflmomwoflonQnmm
nwd

n @5853 98 5183
mod u 2QO ES

Adam?

9:03:88 5:036:00
3633 5:323

omé ”$88, mofiomnzm
omé “owns“ 800m 3on

$3
28988 98 308

_afitamhmomwofinowmm

@6883 2a 532
62:2:
mEmEOU .Som .GOmuOQ on“

8 “comma of mo 883895

 

05 ES om: mo 300me
83.538 @855“ Boa 2335 got? an? souoflmumm
n25:80 $223» 3285me wESom Eon $83 :2: 2:0: vM £5 mo bzmso ooggompom
moEoanozmm wcfloom noumtomom 8:982 03mg;

 

/l

9035259» N DENH

42

blood components. This variable was measured with the Chemistry Abnormality Score
(CAS) (Wolcott & Nissenson, 1988).

The CAS, which was calculated from the results of 11 biochemical tests drawn
before dialysis and during the month of participation in the study, provided a summary of
ESRD patients’ physiological status (Wolcott & Nissenson, 1988). The tool used to
compute the CAS includes a listing of the 13 laboratory tests and values that are
considered normal, mildly abnormal, and abnormal. Hematocrit, serum glutamic-pyruvic
transaminase (SGPT), serum glutamic-oxaloacetic transaminase (SGOT), alkaline
phosphotase, total protein, albumin, triglycerides, carbon dioxide, calcium, and
phosphorus are scored as normal (0), mildly abnormal (1), or abnormal (2). Three of the
variables (i.e., potassium, blood urea nitrogen, creatinine) are scored as normal = O or
abnormal = 1. The potential range of CAS scores is O — 23, with higher scores indicative
of more abnormal physiological status. Because triglyceride and SGPT levels are not
routinely assessed in patients treated at the two clinics used for data collection, the
possible range of CAS scores for this study was 1 — 19.

In developing the CAS, the investigators used a modified Delphi approach to
determine which laboratory values to include in the scoring as well as the criterion values
and the cut-off points for abnormality in the ESRD population (A. Nissenson, personal
C0Inlnunication, April 14, 2003). The Delphi method was first developed by scientists at

the Rand Corporation as an iterative, consensus building process for predicting future
eVents. The strategy has since been applied in a variety of fields to develop consensus and
make group decisions (Custer, Scarcella, & Stewart, 1999).

The Delphi technique is described as an exercise in group communication among

43

 

geographically dispersed experts in a field who share the same interests. Three key
elements of the process include: a structured information flow, feedback to the
participants, and participant anonymity (Nelson, 2002). In a modified Delphi approach,
the process begins with a set of items carefully selected from the literature and expert
opinion. In subsequent rounds, the experts suggest modifications based on feedback from
the previous round. Through a series of rounds the process is designed to yield consensus
(Custer et al., 1999).

Laboratory tests and criterion values included in calculation of the CAS were
chosen by consensus as described above. In addition, the laboratory tests assessed are
those routinely performed on hemodialysis patients monthly. Both the method of
development and use of commonly assessed laboratory tests contribute to the content
validity of the CAS (Wolcott & Nissenson, 1988). Internal consistency reliability of the
CAS was not assessed in this or previous studies because doing so was not warranted
conceptually.

However, the accuracy and precision with which biochemical tests are performed
must be addressed. Laboratories that run the monthly blood studies for the dialysis clinics
involved in this study are fully accredited by the College of American Pathologists (CAP)
(B. DeVries, personal communication, January 8, 2004). The CAP is approved as an
accrediting organization for clinical laboratories under the Clinical Laboratory

Improvement Amendments of 1988 (CLIA) program (CLIA Program; Approval of the
College of American Pathologists, 1995). The CAP has established performance
Standards related to all areas of laboratory functioning that must be met for accreditation.

Accreditation acknowledges adherence to these standards (CAP, 2004).

44

The presence of comorbid conditions was also included as a component of disease
severity in the ESRD population. Comorbidity was defined as a significant concurrent
disease in addition to ESRD which involves organs other than the kidneys but which may
also be responsible for kidney failure (Khan 1998).

Comorbid conditions were measured using a demographic questionnaire that
included a list of commonly occurring concurrent diseases. The conditions chosen for the
list were adapted from a comorbidity questionnaire developed by Katz, Chang, Sangha,
Fossel, and Bates (1996). Katz et al. modeled their measure on the Charlson Comorbidity
Index, an extensively validated chart review instrument (Charlson, Pompei, Alex, &
MacKenzie, 1987; Katz et al., 1996). In addition, several of the comorbidities inventoried
in this study are also requested by the Medical Evidence Report Form which is completed
by dialysis facilities at the initiation of dialysis (U SDI-IHS, 2004).

However, hypertension was notably absent from the tool developed by Katz et al.
(1996). Along with diabetes, high blood pressure is one of the most common
complicating conditions present in new dialysis patients (U SRDS, 2001). Hypertension

was added to the list for this reason. The final list presented to patients included the
following comorbidities: heart attack, heart failure, clogged arteries in the legs, numbness
in feet/legs, stroke, chronic lung disease, stomach ulcer/reflux disease, diabetes, arthritis,
lupus, fibromyalgia, cirrhosis of the liver, cancer, and hypertension.

Patients were asked which of the conditions on the list they had been told by a
health professional that they currently experience. Data about comorbidities were
collected via self-report because this method has been found to be valid, reliable, and to

correlate with more expensive chart audits (Katz et al., 1996). Patients were also allowed

45

 

 

to report comorbidities in addition to those listed. A comorbidity score was calculated by
summing the total number of conditions reported from the original list. Additions made
by patients to the list were not included in the scoring.

Dialysis adequacy. Dialysis adequacy was a second major physiological
influencing factor included in this study. It was defined as the effectiveness with which
dialysis treatment removes waste products from the body, specifically urea (National
Institute of Diabetes and Digestive and Kidney Disases [NIDDK], 2001). Dialysis
adequacy was measured using the most recently reported Kt/V (i.e., dialyzer clearance of
urea x time / volume of body water) levels found in patients’ charts. Kt/V is determined
by comparing levels of urea in blood drawn at the beginning and at the end of a dialysis
session. Typically, Kt/V is evaluated monthly. A minimum value of 1.2 is desired (NKF,
1997).

Influencing Factors: Psychological
Anxiety and depression represented the psychological influencing factors in this
study. Anxiety was defined as a state in which the individual experiences feelings of
uneasiness and activation of the autonomic nervous system in response to a vague,
nonspeCific threat (Carpenito, 1997). Depression was defined as the loss of pleasure
response (i.e., anhedonia); namely, the state of reduced ability to experience pleasure
(Snaith, 2003).
The Hospital Anxiety and Depression Scale (HADS) was used to measure the
presence and severity of both anxiety and depression. The HADS was originally used in
the hospital setting with patients 16 to 65 years of age as a screening tool for the

detection of clinically significant anxiety and depression (Zigmond & Snaith, 1983).

46

Subsequent studies have shown the scale to be valid for use in outpatient and community
sites as well as with the elderly (Aylard, Gooding, McKenna, & Snaith, 1987; Flint &
Rifat, 2002; Snaith & Zigmond, 1994; Watts et al., 2002). Because the HADS was
developed for use with physically ill patients, scale items that may be symptomatic of
both physical illness and anxiety and depression were excluded. This makes the tool
especially useful for assessment of ESRD patients who manifest symptoms that might
give misleading scores on other anxiety and depression scales. A number of researchers
have used the tool to assess mood disturbances in the ESRD population for this reason
(Killingworth & Van Den Akker, 1996; Lye, Chan, Leong, & van der Straaten, 1997;
McCann & Boore, 2000).

The HADS is comprised of a total of 14 items; 7 evaluate anxiety and 7 assess
depression. Patients respond using a 4-point numeric rating depending on the severity of
the problem described in each question. Response bias is avoided by alternating the order
of responses so that for some of the items the first response indicates maximum severity
and in others it is the last. In addition, four response choices are included as options to
prevent patients from consistently choosing the middle choice. Items on each of the

subscales are summed, with possible scores ranging from 0-21. A score below 8 is in the
normal range, 8-10 suggests of the presence of the respective state, and greater than 10
indicates probable disorder (Zigmond & Snaith, 1983).

Construct validity was assessed during instrument development by comparing the
HADS scores with a formal psychiatric assessment. Correlations between the psychiatric
evaluations and the subscale scores were high for both anxiety (r = 0.70, p < 0.001) and

depression (r = 0.74, p < 0.001) (Zigmond & Snaith, 1983). The HADS was also

47

 

validated against the Irritability-Depression-Anxiety scale, a tool used to measure these
mood disorders in a variety of settings (Aylard et al., 1987). In addition, an exploratory
factor analysis of the HADS was carried outin a study of 568 patients with cancer
(Moorey, et al., 1991). Two distinct but correlated factors emerged that corresponded to
the questionnaire’s anxiety and depression subscales.

Internal consistency reliability was examined in preliminary studies of the tool
(Zigmond & Snaith, 1983). For the anxiety subscale, inter-item correlations ranged from
0.41 to 0.76. The items in the depression scale had correlations ranging from 0.30 to 0.60.
Further assessment of reliability by Moorey et a1. (1991) revealed a Cronbach’s alpha of
0.93 for the anxiety scale and 0.90 for the depression scale. For this study, evaluation of
the internal consistency reliability for the full scale resulted in a Cronbach’s alpha of 0.87
(n = 129). A Cronbach’s alpha of 0.84 was found for the anxiety subscale and 0.79 for the
depression subscale.

Influencing F actors: Situational/Demographic

Situational/demographic influencing factors referred to aspects of the social and
physical environment that may affect symptoms including: age, gender, race, marital
status, level of education, and length of time on dialysis (Lenz et al., 1997). These
variables were measured using a demographic questionnaire designed by the investigator.

Age was reported in years and length of time on dialysis was reported in months.
Racial categories included: African American or Black, Hispanic or Latino, Native
American or American Indian, Asian or Pacific Islander, White non Hispanic, or other.
Married, living with husband/wife; single, never married; divorced/separated; or

widowed were Options available for marital status. Level of education was ranked as 8th

48

 

 

grade or less, some high school, high school diploma or GED, vocational school or some
college, college degree, or professional/graduate degree.
Symptoms

Symptoms were defined for this study as multidimensional phenomena

characterized by their occurrence, distress, and concurrence. Occurrence included the
symptom’s severity (intensity), frequency (how often the event occurs within a given
time frame), and duration (how long it lasts) (McDaniel & Rhodes, 1995). The degree of
bother or physical/emotional upset associated with a symptom defined distress (Rhodes
& Watson, 1987). The final symptom dimension, concurrence described the coexistence
and interaction of multiple symptoms (Lenz et al., 1997).

All symptom dimensions were measured using a disease specific physical
symptom tool developed for ESRD patients by Parfrey et a1. (1988; 1989). This tool was
chosen because it taps several symptom dimensions. However, the original tool was
adapted in consultation with Parfrey (personal communication, April 29, 2003) to include
all symptom dimensions of interest to the current study. Since distress was not assessed
in the original instrument, items that questioned the impact of the symptom on QOL and
activities of daily living were changed to assess this dimension. In addition, response sets
for the duration and frequency items were changed fiom yes/no ratings to an ordinal
scale. The current form of the tool assesses the severity, frequency, duration, and distress
associated with the same 11 symptoms included in the earlier tool developed by Parfrey
and colleagues (1988) and commonly experienced by hemodialysis patients. The physical
symptoms assessed included: tiredness, itching, headaches, problems sleeping, joint pain,

cramps, shortness of breath, chest pain, nausea and/or vomiting, abdominal pain, and

49

 

 

muscle weakness. An ‘other’ category is also included for reporting symptoms
experienced, but not among those listed.
Patients were asked to rate each symptom dimension using a 5-point numeric
rating scale; namely, severity (l = not at all severe to 5 = very severe), frequency
(1 = not ofien to 5 = every day), duration (1 = not long to 5 = all day), and distress
(1 = a little to 5 = very severely). An aggregate score was calculated for each symptom
by summing ratings on each of the symptom dimensions. The aggregate symptom score
was 0 if the patient did not report the symptom and ranged from 4 to 20 if the patient did
report the symptom. Higher scores correspond to higher levels of the symptom.
Content validity was established for the original tool developed by Parfrey et al.
(1988; 1989) through patient interviews, literature review, and expert opinion. Since the
symptoms included in the adapted tool are identical to those in the original tool, it has
content validity. In addition to Parfrey, a draft of the tool was reviewed by two nurses
who are experts in the field of nephrology nursing. Anita Molzahn is a well known
researcher specializing in QOL issues. She is a professor in the School of Nursing as well
as Dean of Human and Social Development at the University of Victoria, British
Columbia. Therese Winslow is a clinician who has worked with both hemodialysis and
renal transplant patients for over 30 years. She is employed at St. Mary’s Mercy Medical,
Grand Rapids, Michigan, which is the regional kidney center for West Michigan. No
significant changes were made in the tool following feedback.
Reliability was evaluated during the course of the study. Approximately every
fifth participant enrolled (n = 24) completed the tool on two occasions one week apart

and on the same day of the week. The total symptom scores for the two administrations

50

 

were highly correlated (r = 0.78, p _<_ 0.01). Likewise, the total number of symptoms
reported the first time was highly correlated with the number reported the second
(r = 0.71, p S 0.01). The instrument demonstrated strong test-retest reliability. Also,
paired sample t-tests demonstrated that there were no significant differences between
symptom scores (t(23) = 0.12, p = 0.91) or total numbers of symptoms (t(23) = -0.54,
p = 0.59) reported at Time 1 and Time 2.
Level of Symptom Relief
Level of symptom relief was defined by three components: the strategies used to
obtain relief from symptoms, the degree to which symptoms were relieved, and the
satisfaction with the degree of relief achieved. Symptom relief strategies were defined as
the methods used by patients to achieve a level of symptom relief. Strategies included
those that were prescribed by healthcare professionals as well as any that were
self-initiated. Degree of relief was defined as a subjective evaluation of the extent to
which symptoms were relieved. Perceived satisfaction with relief was the extent to which
the individual was content with current level of symptom relief.

A review of the literature revealed no existing tool to measure either the level of
symptom relief or the relief strategies used by hemodialysis patients. Therefore, an
instrument was developed specifically for the study to assess these variables. The tool
was designed to correspond to the symptom measure previously described in order to
assess the level of relief obtained from each of the reported symptoms.

The instrument is comprised of two parts: strategies used to relieve symptoms and
the level of relief achieved. Part 1 of the tool lists strategies commonly used to relieve

each of the symptoms, including those prescribed by healthcare providers as well as

51

 

 

self-initiated methods. Patients are asked to indicate which of the listed strategies are
used as well as the one that they perceive as the most effective in relieving each of the
symptoms. An ‘other’ category is included for patients to report strategies used in
addition to those listed. This part of the tool obtains descriptive data regarding methods
of symptom relief.

Part 2 of the tool evaluates the degree of relief obtained using the methods
identified in Part 1. Patients respond to a general question regarding their level of relief
using the following 6-point numeric rating scale: 0 = no relief to 5 = complete relief.
Patients are also questioned as to their degree of satisfaction with the relief they obtain
from symptoms. Responses range from 0 to 5; ‘not at all satisfied’ to ‘completely
satisfied.’ In addition, patients are asked to indicate whether their actions decreased each
dimension individually using a 6-point numeric scale to rate each dimension; 0 (not at all)
to 5 (very much).

The tool results in three scores for each of the 11 symptoms: an aggregate
Symptom relief score, a relief subscale score, and a satisfaction subscale score. A total
relief score is calculated for each symptom by adding: (a) the rating given on the general
relief question, (b) the rating on the satisfaction with relief question, and (c) the ratings
given for relief relative to each of the symptom dimensions. Possible scores range from 0
to 30. The satisfaction subscale score is the rating given on the satisfaction item only
(range = 0-5). A relief subscale score is calculated from all items minus the satisfaction
question (range = 0-25). In all cases, higher scores are indicative of a higher level of
relief achieved.

Content validity is supported by the literature of common symptom management

52

 

techniques (Ferrell & Coyle, 2001). This measure was also reviewed by the same experts
in nephrology nursing who reviewed the symptom tool. As with the symptom tool, 24
patients completed this instrument twice, one week apart on the same dialysis day.
Test-retest reliability was only moderate (r = 0.47, p _<_ 0.05). However, there was not a
significant difference found between total symptom relief scores at Time 1 and Time 2
(t(23) = O.18,p = 0.86).
Performance: Quality of Life
The consequence of interest to this study was the QOL of hemodialysis patients.

QOL was defined as a person’s sense of well-being that stems from satisfaction or
dissatisfaction with the areas of life that are important to him/her (Ferrans & Powers,
1992, p. 29).

The Ferrans and Powers (1985) Quality of Life Index—Dialysis Version (QLI)
was used to measure QOL. The QLI is a two part, self-report questionnaire that assesses
both satisfaction with various aspects of life and the importance of each aspect of life to
the person (Ferrans & Powers, 1985). Each part is composed of a total of 32 items that
evaluate the following four domains: health and functioning, psychological/spiritual,
social and economic, and family. In addition, the hemodialysis version includes two
additional items related to dialysis treatment.

Participants respond to statements using a 6-point numeric rating scale ranging
from ‘very dissatisfied’ to ‘very satisfied’ for the satisfaction items and from ‘very
unimportant’ to ‘very important’ for the importance items. Scoring the QLI involves
centering and weighting of scores. The step-by—step scoring instructions, which include a

computer program written for use with SPSS, were used to score the QLI in this study.

53

 

 

These instructions are available fi'om the authors on their website (Ferrans, 2004). The
tool yields a total score as well as four subscale scores corresponding to the domains of
life assessed. Potential scores for both the subscales and total score range from 0-30, with
higher scores indicative of higher QOL.
An advantage of the QLI is that it is constructed in a manner that prevents
problems with collinearity among the study variables. Unlike other QOL instruments, the
QLI does not include a symptom scale. Energy is the only symptom that is assessed using
two items. One item questions the importance of having enough energy for everyday
activities. A second asks about the level of satisfaction with the amount of energy
actually present. It does not evaluate symptom dimensions.
Similarly, although the patients’ marital status is examined on both the
demographic questionnaire and the QLI, the questions on each tap different concepts.
Patients are asked to indicate which marital status category they belong to on the
demographic questionnaire. In contrast, questions related to marital status included on the
QLI assess the status of the relationship not the category to which the patient belongs. To
validate the absence of collinearity, the tolerance associated with marital status was
examined when QLI scores were regressed on the demographic/situational influencing
factors. A tolerance of 0.84 indicated that marital status was independent of other
variables (George & Mallery, 2003).

The validity of the QLI has been established. Content validity has been supported
by the fact that items included on the tool were based on an extensive review of the QOL
literature as well as patients’ ideas regarding the characteristics of a quality life (F errans

& Powers, 1985). Further support for content validity was provided by an acceptably

54

 

high rating using the Content Validity Index (Oleson, 1990).

Construct validity of the QLI was established through factor analysis. The factor
analysis solution, which revealed the four domains of QOL listed previously, accounted
for 9 1 % of the total variance. Factor analysis of the four primary factors also revealed
one higher order factor, representing QOL (Ferrans & Powers, 1992). The contrasted
group approach was also used to support construct validity. Participants were divided into
groups based on self-reported levels of pain, depression, and coping skills. Those who

were more depressed, exhibited higher levels of pain, and coping less well also reported
lower QOL (F errans, 1990). The QLI was also compared with other similar instruments
to establish convergent validity. For example, strong correlations were found between the
total QLI score and Campbell, Converse, and Rodger’s measure of life satisfaction
(Ferrans & Powers, 1985).

Both internal consistency and temporal reliability have been established. Internal
consistency reliability has been evaluated with several groups. However, of interest to
this study were findings specific to the dialysis population. Cronbach’s alphas for the
total QLI in studies with dialysis patients have ranged from 0.90 to 0.93. Internal
consistency reliability was established for the subscales in one study with dialysis
patients with the following results: health and functioning 0.87, psychological/spiritual
0.90, Social and economic 0.82, and family 0.77 (Ferrans & Powers, 1992). Test-retest

reliability has also been demonstrated in studies that spanned a 2 to 4 week interval
(F errans & Powers, 1985).
The tool was also found to be sensitive to change, with QLI scores changing

significantly over time following an intervention (Ferrans & Powers, 1992). This

55

 

property will be of importance to future studies that evaluate the impact of interventions
on symptom experience and QOL. Finally, this tool was an appropriate QOL measure
because it was written at a fourth grade level and has been tested for use with racial and
cultural groups sampled for this study (F errans, 1997).
Internal consistency reliability of the QLI was evaluated for this study. The
Cronbach’s alpha for the entire QLI was 0.93. Cronbach’s alphas were also computed for
the subscales with the following results: health and firnctioning 0.87,
psychological/spiritual 0.91, social and economic 0.75, and family 0.61. Reliabilities
were assessed with the full sample of 130 patients. The moderate value noted for the
family subscale is comparable with previous reports that have consistently noted the
lowest alpha for this subscale (Deshotels, Planchock, Dech, & Prevost, 1995; Ferrans,
1990; Ferrans & Powers, 1992).
Procedures
The study was approved by the Michigan State University Committee for

Research Involving Human Subjects (UCRIHS) as well as the review boards of both
dialysis clinics. In light of HIPPA regulations enacted prior to the beginning of data
collection, no contact with patients was allowed without prior written consent. Therefore,
staff nurses in both clinics were required to obtain a pre-consent using a form approved
by UCI{IHS and both clinic review boards. Registered nurses and technicians from each
of the dialysis units were introduced to the study as well as their role in obtaining
PTC‘COHSents at a meeting held prior to beginning of data collection. Following this
orientation, nurses approached patients as time allowed and explained only that a

research study was being conducted and that the investigator was requesting permission

56

 

to introduce herself and explain the study. Patients were assured that signing the
pre-consent did not obligate them to participate once the study had been explained.

Once pre-consent was obtained, each patient was approached individually,
introduced to the study, and asked to participate if he/she met the inclusion criteria. Of
patients who signed pre-consent forms, only three individuals elected not to participate
after the study was explained. These patients stated that they were either too tired to
participate or were not interested. Written informed consent was obtained from those who
agreed to participate after the study was explained. Percentage of participation from each
clinic was moderate; 48% of the patients treated at Clinic A and 44% of those receiving
dialysis at Clinic B completed interviews. Due to privacy laws, no information was
available about those patients who declined to sign pre-consent forms or agree to

Participate in the study.

Patients were interviewed immediately after obtaining informed consent.
Although patients were given the option of completing the interview in a private room
rather than in their dialysis chairs in the open unit, none desired to do so. Hemodialysis
sessions typically required from 3 to 4 hours. Only approximately the first 1/2 hour and
last ‘/2 hour were occupied with activities involved in initiating and completing treatment.
Interviewers were not allowed in the vicinity of the dialysis chairs during those times.
9111:1118 the remaining time, however, patients were free to engage in sedentary activities

such as talking.

All questionnaires were administered via interview to maintain uniformity of

administration. This was a necessity since the majority of patients found it difficult to

write due the need to keep their arms still for dialysis and blood pressure monitoring.

57

 

Many were also not able to see or write well enough to complete the measures unaided.
Patients were interviewed on three shifts, usually during the periods from 8:00

am. to 11:30 am, 12:30 pm. to 3:30 pm, and 5:00 pm. to 8:00 pm. Interviews
required from 1/2 hour to 2 hours, with an average of 45 minutes, depending on the
number of symptoms patients reported. Only one interview was not completed in one
session due to a hypotensive episode experienced during the dialysis treatment. All of the
questionnaires but the QLI were finished prior to the onset of low blood pressure, at
which time the interview was terminated. The QLI was completed during the next
dialysis session. No further problems arose during completion of the interview. No
patients refused or failed to complete the entire interview. No patients experienced undue
fatigue during the interview and many expressed appreciation for the chance to talk about
their symptoms.

Interviews were conducted by the primary investigator (76%) and two research
assistants (24%). Upon hire, research assistants were oriented to the purpose of the study,
privacy issues, and their job responsibilities. All interviewers also completed the tutorial
provided by the National Institute of Health regarding the rights of human subjects. They
observed from one to three interviews prior to conducting their first. They were also

instructed in completing the CAS. Once data were entered into the computer database,
one Of the research assistants rechecked all entries for accuracy.
Protection of Human Subjects

Additional precautions were taken to protect participants’ confidentiality during

the COurse of the study. All information obtained from the patients and their medical

records was identified by respondent number only. No names appeared on any of the

58

 

questionnaires completed by patients. A master list of participants and their identification
numbers is being kept on a disc in a locked disc box. All consent forms are being stored
in a locked file separate from the interview questionnaires. All participant files are being
stored in a fire-proof locked file cabinet. There is no link between the master list and the
database. All data generated from this study will be kept in a locked file cabinet for a
period of three years following completion of the study. A copy of the original database
as well as information regarding location of the data will be kept on file in the College of
Nursing Research and Doctoral Program office at Michigan State University for that
period of time as well.

Data Analysis

Sample demographics. Demographic characteristics of the sample were analyzed
using descriptive statistics including frequencies, measures of central tendency, and
measures of variability. This analysis included age, gender, race, marital status, level of
education, length of time on dialysis as well as physical and mental health characteristics
of the sample.

Research questions. The research questions were analyzed in the following
manner:

Question 1: What physiological, psychological, and situational/demographic
influencing factors predict the level of symptoms experienced by ESRD hemodialysis
patients?

Analysis: Regression analysis was employed to determine which influencing
factors predicted symptom scores. Each category of influencing factors was entered into

the analysis beginning with situational/demographic variables followed by psychological

59

and physiological. Nonsignificant variables with p values 0.50 were trimmed from the
model at each step of the analysis.

Question 2. What are the characteristics of symptoms reported by ESRD
hemodialysis patients in relation to the following symptom dimensions: symptom
occurrence (i.e., severity, frequency, duration), symptom distress, and symptom
concurrence?

Analysis: Frequency distributions and descriptive statistics were employed to
examine the symptom dimensions of severity, frequency, and duration. An exploratory
factor analysis was performed to analyze symptom concurrence.

Question 3. What strategies are used by ESRD hemodialysis patients to relieve
symptoms?

Analysis: The methods used by patients to relieve their symptoms were examined
using descriptive statistics: frequencies and percentages.

Question 4. What level of symptom relief is achieved by ESRD hemodialysis
patients?

Analysis: Descriptive statistics were employed to answer this research question,
including measures of central tendency and variability.

Question 5. What is the relationship between influencing factors and symptoms
and the QOL reported by ESRD hemodialysis patients?

Analysis: Correlation and regression analysis was used to explore the relationship
between influencing factors and symptoms and QOL. A series of regression analyses
were conducted to determine significant predictors of QOL. Each category of influencing

factors was entered into the regression beginning with situational/demographic followed

60

by psychological and physiological. Again, nonsignificant variables with p values 0.50
were removed from the model at each step. Symptom scores were added to the regression
analysis after influencing factors.

Question 6. What is the moderating effect of level of symptom relief on the
relationship between symptoms and QOL?

Analysis: This research question was analyzed using regression analysis
including: all variables retained in the model following analysis of Question 5, level of
relief, and the interaction of symptoms and level of relief. As Aiken and West (1991)
suggest, both symptoms and level of symptom relief were centered around group means.
The moderating effect of level of relief on the relationship between symptoms and QOL

was tested with the interaction term.

61

Chapter 4
RESULTS

Descriptive statistics for each component of the study model will be presented
first, followed by results that address each of the research questions. Descriptive statistics
for level of relief will be included with the discussion of the research question that
addresses that component of the model. All analysis were done using SPSS 10.0 version.
Influencing Factors: Physiological

Disease severity. The hemodialysis patients in this sample maintained adequate
physiologic balance with the exception of phosphorus (M = 5.74 mg/dl, SD = 1.75) and
alkaline phosphatase (M= 122.37 U/L, SD = 77.10). See Table 3 for results of all laboratory
tests and acceptable levels for ESRD patients (Good Samaritan Hospital Chronic Dialysis
Center, 2004). Further evidence of biochemical stability was provided by low CAS scores,
M = 1.84, SD = 1.49 (possible range 0-19).

Patients reported an average of four comorbid conditions. High blood pressure was
by far the most prevalent concurrent illness, reported by 82% of the sample. Peripheral
neuropathy (49%), stomach ulcers/reflux (48%), arthritis (46%), and diabetes (46%)
rounded out the top five comorbidites in descending order of occurrence. See Table 4 for
the frequency of all 14 comorbid conditions listed on the demographic questionnaire.

Dialysis adequacy. Patients also received an adequate dose of dialysis according to
the National Kidney Foundation (NKF) (1997) Clinical Practice Guidelines. Mean Kt/V
(dialyzer clearance of urea x time / volume of body water) values were above the level
recommended (M = 1.55 ml/min, SD = 0.35). The established benchmark for dialysis

adequacy is a Kt/V 3 1.2 ml/min (NKF, 1997). Values at or above this standard

62

 

Table 3

Physical and Mental Health Characteristics of Sample

 

Acceptable
Characteristic M (SD) Normal value ESRD normal
Laboratory values (pre-dialysis)
Potassium (meq/L) 4.89 (0.63) 3.5-5.4 4.0-6.5
Creatinine (mg/dl)—male 9.39 (2.80) 0.5-1 .1 10-20
Creatinine (mg/dly—female 8.58 (2.74) 06-12 10-20
Blood urea nitrogen (mg/d1) 60.47 (18.08) 7-25 40-85
Hematocrit (%)—male 37.10 (3.74) 40-54 30-36
Hematocrit (%)—female 37.36 (4.56) 37—47 30-36
Alkaline phosphatase (U/L) 122.37(77.10) 30-1 15 30-1 15
Albumin (g/dl) 3.84 (0.40) 3.5-5.5 3.8
Calcium (mg/d1) 9.22 (0.89) 85-108 85-108
Phosphorus (mg/d1) 5.74 (1.75) 2.5-4.5 3.5-5.5
CAS 1.84 (1.49) NA NA
Number of comorbid conditions 4.29 (1.95) NA NA
Dialysis adequacy (Kt/V) 1.55 (0.35) NA 1.2
HADS—anxiety 5.95 (4.76) 7 7
HADS—depression 5.66 (4.03) 7 7

 

Note. NA = not applicable; HADS = Hospital Anxiety and Depression Scale;

CAS = Chemistry Abnormality Score; Kt/V = dialyzer clearance x time / volume of body

water (approximately 60% of body weight).

63

Table 4

Prevalence of Self-Reported Comorbid Conditions

 

Condition n %
High blood pressure 106 82
Numbness in feet/legs 63 49
Stomach ulcer/reflux disease 62 48
Arthritis 60 46
Diabetes 60 46
Heart failure 55 42
Heart attack 45 35
Clogged arteries in legs 32 25
Cancer 27 21
Chronic lung disease 22 17
Stroke 18 14
Lupus 5 4

F ibromyal gia 2 2

Cirrhosis 0 0

 

64

meet the NKF guidelines for adequacy.
Influencing Factors: Psychological

Anxiety and depression. Scores on the anxiety subscale of the HADS varied
considerably (range = 0-21). However, the average anxiety score was within the normal
range (M = 5.95, SD = 4.76). Depression scores also varied across individuals
(range = 0-19). The mean depression score was 5.66 (SD = 4.03), also within normal
limits.

Breaking down the anxiety scores by category, 68% of the patients’ scores fell in
the normal range (0-7), 12% in the range suggestive of the disorder (8-10), and 20% in the
probable range ( 3 11). Depression scores were categorized as follows: 70% normal (0-7),
18% possible presence of the disorder (8-10), and 12% probable presence of depression
(_>_ 11)-

Influencing Factors: Situational/demographic

Age. Patients ranged in age from 22 to 88 years of age, with a mean of 60.22 years
(SD = 32.25).

Gender. The sample was evenly divided by male and female.

Race. Racial composition of the full sample was 69% White, non Hispanic; 27%
Black; and 4% other racial minorities. When the two dialysis clinics were examined
separately, the racial make-up of the sample from Clinic A was 84% White and 12% Black.
The sample from Clinic B included a greater percentage of minority patients (i.e., 53%
Black, 43% White).

Marital status. The majority of patients were married, living with a spouse (52%).

The second largest category included patients who were divorced or separated (24%). The

65

final 24% was comprised of individuals who were widowed or had never been married.

Level of education. Educational achievement varied, but 72% of the sample
graduated from high school at a minimum. Only 8% had eighth grade or less education.

Length of time on dialysis. Mean and median length of time on hemodialysis were
32.25 months and 20 months, respectively. There was wide variability noted in treatment
duration, with a range of 0.42 months to 20 years. Fifty-nine percent of the patients had
been on dialysis for 2 years or less, only 3% had been treated for more than 10 years.
Symptoms

Patients reported an average of 5.67 symptoms. Tiredness, the most prevalent of
the 11 physical symptoms assessed, was experienced by 77% of the sample. Following
close behind were problems sleeping, which were report by 63% of the patients. Least
commonly occurring was chest pain. In addition to those listed, two symptoms frequently
reported in the ‘other’ category were numbness/tingling in the hands and feet (49%) and
restless legs (22%). See Table 5 for the prevalence of all 11 symptoms assessed in
addition to others not listed but reported by at least 10 patients.
Level of Relief

Descriptive statistics for level of relief will be discussed in relation to research
question number 4.
Quality of Life

Total scores for the QLI ranged from 7.02 to 29.22, with a mean of 20.76
(SD = 5.06). The mean score for the family subscale was the highest at 24.68
(SD = 5.59), followed by the psychological/spiritual subscale (M = 22.40, SD = 6.55) and

the social and economic subscale (M = 22.17, SD = 5.35). Mean scores for the health and

66

 

Table 5

Prevalence of Symptoms Reported by ESRD Hemodialysis Patients

 

Symptom n %
Tiredness 100 77
Problems sleeping 82 63
Cramps 68 52
Muscle weakness 67 51
Joint pain 63 48
Itching ‘ 56 43
Shortness of breath 43 33
Nausea / vomiting 41 32
Headaches 39 30
Abdominal pain 18 14
Chest pain 17 13
Others*

Numbness / tingling in hands, feet 64 49
Restless legs 29 22
Changes in taste / smell 14 11
Decreased appetite 12 9

Other pain 10 8

 

Note. * Symptoms reported in the “other” category of the symptom tool by at

least 10 patients

67

 

 

functioning subscale were the lowest (M = 17.88, SD = 5.89). The possible range of
scores is 0-30 for the total QLI and subscales, with higher scores indicative of better
QOL.

Research Questions

The remaining results of the study are presented according to the research questions
posed.

Question I: What physiological, psychological, and situational/demographic
influencing factors predict the level of symptoms experienced by ESRD hemodialysis
patients? Prior to conducting regression analysis, a total symptom score was calculated
for each participant by summing the scores received for all of the symptoms reported by
the individual. This value was used in all regression analyses with symptoms as the
dependent variable. Correlations between predictor variables and symptom scores were
also calculated. A series of regression analyses were then conducted to evaluate which of
the influencing factors included in the study model were significant predictors of
symptoms.

The influence of situational/demographic influencing factors (i.e., age, gender,
race, marital status, level of education, and length of time on dialysis) on symptom scores
was examined first. The correlations among these predictor variables were generally low
(see Table 6). The only significant correlations were seen between age and race
(r = -0.32, p < 0.01), age and level of education (r = -0.32, p < 0.01), and race and marital
status(r = -0.20, p < 0.05). A greater number of older patients were White. Older
participants also had less education. White patients were more often married than

non-White patients. Age was the only situational/demographic predictor that was

68

 

 

 

|ll
_00 v a r. .80 v a ._. .202
>233?

-- 8305
8.0- -- 0385
00.0 m a .0 -- use? 03
2.0- :0 ._.:.0 -- ause
8.0- 8.0 1.3.0 :80 -- 583.

8520
8.0 :0- 00.0 00.0- 2.0- -- 8 85
:0- 00.0- 8.0 8.0- 00.0 8.0- -- 000m
20. ...:.0- .08- 02.0- 3.0- 8.0- 2.0 -- mean a:
00.0 :0- 5.0- 00.0- 8.0 s a .0 8.0- 88. -- 8%
3.8.0 8.0 8.0 2.0- 2.0 00.0- 00.0 2.0- 8.0 -- 8080
:0 8.0 3.8.0- ._.:.0- :80. 00.0- 3.8.0. 3.0 1.8.0- 8.0- -- owe.
596200 32:38 mofig mama—2e £530
£335 #058800 83 mmoaom booze no 08E- ozem amt-82 comm economy owe-

 

Eocoek meaotoébé NeowmoNowwém use .Neommgoeokwm GSQEMQEmmSeeoteENW MaoEV weonfimtob

o 033-

69

significantly correlated with symptom scores (r = -0.29, p < 0.01), indicating that older
patients had lower symptom scores than younger dialysis patients.

Symptom scores were regressed on all situational/demographic influencing
factors. See Table 7 for a summary of regression 1 with symptom score as the dependent
variable. Analysis revealed R2 = 0.17, F(6,123) = 4.06, MSE = 1088.22, p = 0.00. Age
and race were significant predictors of symptoms. Marital status and length of time on
dialysis were trimmed from the model because they did not contribute any unique
variance to symptoms and were nonsignificant at a p _>_ 0.50. Gender and level of
education were retained in the model due to parameter estimates that neared significance.

Psychological influencing factors were added to the model in the next step of the
analysis. Correlations among the situational/demographic predictor variables, anxiety,
and depression are provided in Table 6. Age was negatively correlated with anxiety
(r = -0.43, p < 0.01) and depression (r = -O.17, p < 0.05) as well as with symptom scores
(r = -0.29, p < 0.01). Reporting less anxiety and depression, older patients also reported
lower levels of symptoms. Not surprisingly, anxiety and depression were moderately and
positively correlated with each other as well as with symptom scores.

See regression 2 on Table 7 for a summary of the regression analysis adding the
psychological influencing factors to the model. Forty-one percent of the variance in
symptoms was accounted for in this step, R2 = 0.41, F (6,122) = 14.23, MSE = 772.57,

p = 0.00. Anxiety and depression contributed significantly to the model, adding 24% of the
variance. Race was also a significant predictor of symptom scores.

The last category of influencing factors was then added to the model. Physiological

factors included Chemistry Abnormality Scores (CAS), comorbid conditions, and adequacy

7O

Table 7

Summary of Regression Equations Identifiang Predictors of Synflotoms

 

 

Regression Independent
variables R2 B SE [3 t p
1 Influencing factors:
situational/
demographic 0. 1 7
Age 092 0.21 -0.41 -4.32 0.000
Gender 9.48 5.90 0.14 1.61 0.111
Race -19.58 6.89 -0.26 -2.84 0.005
Marital status -2.55 6.06 -0.04 -0.42 0.675
Level of education -5. 10 2.97 -0. 15 -1.72 0.089
Time on dialysis 0.01 0.09 0.01 0.12 0.904
2 Influencing factors:
trimmed
situational/
demographic,
psychological 0.41
Age -0.37 0.20 -0.16 -1.82 0.071
Gender 9.46 5.15 0.13 1.84 0.069
Race -12.14 5.88 -0.16 -2.06 0.041
Level of education -2.46 2.55 -0.07 -0.97 0.337
Anxiety 2.47 0.70 0.33 3.56 0.001
Depression 2.54 0.75 0.29 3.38 0.001
3 Influencing factors:
trimmed situational/
demographic,
psychological,
physiological 0.50
Age -0.28 0.20 -0.12 -1.37 0.175
Gender 5.42 5.35 0.08 1.01 0.314
Race -6.31 5.84 -0.08 —1.08 0.283
Level of education 0.22 2.55 0.01 0.09 0.930
Anxiety 2.83 0.69 0.38 4.09 0.000
Depression 1.79 0.76 0.21 2.36 0.020
CAS -0.07 1.83 -0.00 -0.04 0.968
Comorbid 5.57 1.28 0.31 4.34 0.000
conditions
Dialysis adequacy -4.79 7.33 -0.05 -0.65 0.515

 

71

 

 

Table 7 (continued)

 

Regression Independent

variables R2 B SE [3 t J)

4 Influencing
factors:
trimmed
situational/
demographic,
psychological,
physiological 0.48

Age

Gender

Race

Anxiety

-0.28
6.68
-8.67
2.81

0.18
4.89
5.44
0.65
0.72

-O.12
0.10
-0.11
0.38
0.22

-l.59
1.37
-1.60
4.31
2.72

0.114
0.175
0.113
0.000
0.008

Depression 1 .96
Comorbid 4.95 1.23 0.27 4.03 0.000
conditions

 

 

72

 

of hemodialysis. Of the physiological influencing factors, CAS was significantly correlated
with age, anxiety, depression, and marital status (see Table 6). Chemistry Abnormality
Scores were lower in older patients but higher in those reporting more anxiety, depression,
and unmarried marital status. Comorbid conditions and marital status were also
significantly correlated. Patients who were not married reported higher numbers of
comorbid conditions. Dialysis adequacy was significantly correlated only with gender.
Average Kt/V values were higher in females, indicating a higher delivered dose of dialysis.

Of the physiologic variables, CAS (r = 0.23, p < 0.01) and numbers of comorbid
conditions (r = 0.34, p < 0.01) were significantly correlated with total symptom scores.
Correlations between all individual laboratory values and physical symptoms were also
computed. Significant correlations were noted between phosphorus levels and itching
(r = 0.27, p < 0.05), shortness of breath (r = 0.32, p < 0.05), and chest pain (r = 0.63,

p < 0.01). Blood urea nitrogen and sleep (r = 0.30, p < 0.01) and serum glutamic-
oxaloacetic transaminase and joint pain (r = 0.26, p < 0.05) were significantly correlated as
well.

At this point in the analysis, physiological influencing factors in combination with
trimmed situational/ demographic and psychological variables accounted for 50% of the
variance in symptom scores (R2 = 0.50, F(9,110) = 12.40, MSE = 671.79, p = 0.00). Refer
to Table 7, regression 3 for the results of this analysis. Anxiety and depression were again
significant predictors of symptom scores. Of the physiological factors, only comorbidity
was significant. Laboratory values, adequacy of dialysis, and level of education were
dropped from the model because they were nonsignificant predictors of symptom scores at

p2050

73

The last regression in this series of analyses was conducted including the
situational/demographic, psychological, and physiological influencing factors that were
retained in the model. The final model accounted for 48% of the variance in symptom
scores (R2 = 0.48, F(6, 122) = 18.53, MSE = 687.14, p = 0.00) (see Table 7, regression 4).
Significant predictors of symptom scores include: anxiety, depression, and numbers of
comorbid conditions. The standardized regression coefficient for anxiety was [3 = 0.38,
t(122) = 4.31,p < 0.00, for depression 0 = 0.22, t(122) = 2.72, p < 0.01, and for comorbid
conditions [3 = 0.27, t(122) = 4.03, p < 0.00. Thus, a person who was one standard
deviation above the mean on anxiety was 0.38 standard deviations above the mean on
symptom score. Likewise, a person who was one standard deviation above the mean on
depression was 0.22 standard deviations above the mean on symptom score. Finally, a
person who was one standard deviation above the mean on number of comorbid
conditions was 0.27 standard deviations above the mean on symptom score. Examination
of tolerance showed no indication of multicollinearity, although values for anxiety (0.56)
and depression (0.63) were somewhat low.

Question 2: What are the characteristics of symptoms reported by ESRD
hemodialysis patients in relation to the following symptom dimensions: symptom
occurrence (i. e., severity, frequency, duration), symptom distress, and symptom
concurrence? Primarily descriptive statistics were used to investigate the
multidimensional characteristics of symptoms reported by patients in this study. Factor
analysis was employed to explore the dimension of concurrence.

Table 8 presents an overall summary of several symptom characteristics. The

average total symptom score was 13.41 , rated on a 0-20 scale. The typical symptom was

74

Table 8

Average Characteristics of Symptoms Reported by Hemodialysis Patients

 

 

Characteristic Range of

possible scores M SD
Number of symptoms reported NA 5.67 2.63
Symptom score 0-20 13.41 2.57
Symptom severity 0-5 3.63 0.68
Symptom frequency 0-5 3.33 0.93
Symptom duration 0-5 2.96 0.96
Symptom distress 0-5 3.50 1.03
Symptom relief score 0-30 14.48 6.73

Note: NA = Not applicable
75

 

 

 

 

moderate to severe in severity (3.63), occurred every other day (3.33), lasted
approximately half of the day (2.96) and was moderately to severely distressing (3.50).
All dimensions were rated on a 0-5 scale. Of the 11 symptoms assessed, muscle weakness
received the highest total symptom score. Scores for joint pain, problems sleeping, and
tiredness followed in descending order. The lowest score was for nausea/vomiting (see
Table 9).

Muscle weakness was also rated as the most frequently occurring symptom
(M = 4.30, SD 1.30) as well as the longest lasting (M = 4.30, SD = 1.33) and the most
distressing (M = 4.03, SD = 1.23). Cramps of all types were rated as the most severe
symptom (M = 4.12, SD = 1.00). However, cramps typically lasted less than a quarter of
the day (M = 2.10, SD = 1.07) and happened only 1-2 times per week (M = 1.66,

SD = 1.19). When they did occur, though, cramps were moderately distressing (M = 3.79,
SD = 1.37). Difficulty sleeping, ranked as the second most prevalent and severe symptom
(M = 4.02, SD = 1.07), occurred with regularity (M = 3.94, SD = 1.22), was long lasting
(M = 3.71, SD = 0.85), and was severely distressing (M = 3.76, SD = 1.51) to patients.
Ratings for severity, frequency, duration, and distress associated with each symptom are
presented in Table 9.

An inspection of the dimension ratings (i.e., severity, frequency, duration, and
distress) for each symptom failed to reveal any obvious trends in the scores. No one
dimension was consistently rated the highest or lowest. Correlations were also calculated
between the means of each of the symptom dimension in an attempt to uncover any
patterns in their relationships (see Table 10). A high correlation was found between

severity and distress (r = 0.71, p < 0.01). Symptoms that were severe were highly

76

.0309; 05 we £052 Sewn: 88:00: 0280 00:00:: .088 :0 E .oMo .5 .m.: : .36. ”08on .«o omen: 20:000m .862

 

77

80.: : 8.0: 50.: 00.0 c a: 0: :0: :0 A0 _. : 00.0 00.0: 00.: 0.00:8? 0 00802
A000: 00.0: 80.: 00.0 5.: 00.0 A00: 00.: 80.: 00.0 00.0: 00.: 0:8 0026
80.0: 00.0: :0: 00.0 :00: 8.: :00: 0:0 80.: 0:0 8:0: 00.: 8:36
A000: 0.0.: :00. : 8.0 :00: 00.0 $0.: 00.0 80.: 00.0 00.0: :0: 502000 :80
60.: 00.0: :0: 00.0 :0: 00.0 000.: 00.0 00.: 00.0 90.0: 00.0: 88008:
A000: 00.: A00: 00.0 8.: 00.0 60.: 00.0 5.: :00 80.0: 00.0: 00:2:
A00: : 00.0: 00. : 00.0 00.: 00.0 :00. : :0 60.: 00.0 60.0: 00.0: 58 8088.0
20.0: 00.0: 90.: 00.0 :00: 00.0 60.: 5.0 30.0: 00.0 A000: 8.0: 0080000.
$0.0: 00. : :0: 00.0 :80: :0 :00: 00.0 :0: 00.0 50.0: 00.0: 0000020 85:80
80.0: :00: :0: 00.0 80.: 00.0 $0.: :0 00.0: 00.0 $0.0: 00.0: s00 0500
80.0: 00.0 :00: 00.0 00.: 00.0 80.: 00.0 $0.: :00 30.0: 00.0: ause-E 28:2
09:00 N30: 0000000 3:00 09:00 0000000 590800
Boom 08:05 manna—Q 5:00:0on buo>om 280
$wa0 :33. .wmfio>< towfioia :uwfio>< towfo>< 8038000 Eco--

 

EoEEAm. seem gokuggmbammko Rio-.0 03:0 «$05300 nonemENQ 503200090 0808.0. 0083508 6329:0-

0 0:00.

Table 10

Correlations Between Symptom Dimensions

 

 

Mean Mean Mean Mean
severity frequengy duration distress

Mean -- 0.11 0.36** 0.77**
severity

Mean -- 0.55** 0.04

frequency

Mean -- 0.32**
duration

Mean --
Distress

 

Note. ** p 5 0.01

78

distressing as well. Greater severity was also associated with longer duration

(r = 0.36, p < 0.01). The dimensions related to the timing of symptoms were also
significantly correlated (r = 0.55, p < 0.01). Those that lasted the longest also occurred
with the greatest regularity. The frequency with which a symptom occurred was not
significantly associated with either its severity or distress, however.

Exploratory factor analysis was used to investigate the remaining symptom
dimension, concurrence. Appropriate use of factor analysis was determined with tests of
sampling adequacy and multivariate normality. A value of 0.70 was obtained for the
Kaiser-Mayer-Olkin (KMO) Measure of Sampling Adequacy. This value fell above the
‘middling’ range as well as the 0.5 minimum acceptable for factorability (Brown, 2001;
George & Mallery, 2003). Bartlett’s Test of Sphericity was 199.60, p g 0.00, indicating
that the data were approximately multivariate normal and did not produce an identity
matrix (George & Mallery, 2003).

Principal components factor analysis with varimax rotation was used to extract the
factors. Three criteria were used to determine number of factors: the Kaiser-Gunman rule,
the scree plot, and theoretical soundness (Nunnally & Bernstein, 1994). Examination of
the scree plot suggested the extraction of three factors. However, four components emerged
with eigenvalues > 1.0, suggesting a four factor solution according to the Kaiser-Guttmann
criterion. After examination of the symptoms that loaded on each factor, it was decided to
retain a four factor solution on theoretical grounds. The symptoms that were grouped
together had a common thread.

The loadings of the symptoms on the four factors are shown in Table 11. Based on

sample sizes of 100 to 200, factor loadings of 0.384 to 0.512 are the suggested minimum

79

 

 

Table 11
Results of Principal Components Factor Analysis with Varimax Rotation
Including Factor Loadings

Factor
Symptom 1 2 3 4

Tiredness 0.78

Sleeping problems 0.69

Muscle weakness 0.59

Shortness of breath 0.82

Chest pain 0.79

Joint pain 0.80
Headaches 0.63

Itching 0.46
Nausea/vomiting 0.8 l

Abdominal pain 0.75

 

80

 

(Stevens, 2002). In this sample of 130, factor loadings ranged from 0.46 to 0.82. One
symptom, cramps, loaded equally on factor 2 (i.e., shortness of breath and chest pain) and
factor 4 (nausea/vomiting and abdonrinal pain). However, since cramps did not correlate
strongly with any of the symptoms on either factor, it was dropped from the analysis. The
factor solution presented was derived after the elimination of cramps.

Three items loaded saliently on Factor 1. Tiredness, sleeping problems, and muscle
weakness were all related to energy/vitality. Factor 2 included chest pain and shortness of
breath, symptoms typically associated with cardiac related problems. A pain/comfort theme
was apparent among the symptoms comprising Factor 3 (i.e., joint pain, headache, and
itching). Nausea/vomiting and abdominal pain loaded on Factor 4, both of which
accompany problems related to the gastrointestinal system.

The four factor solution accounted for 62.49% of the variance in the model. Fit of
the factor structure was assessed by generating a reproduced correlation matrix. Small
residuals indicated that the model was accurate in its initial explanation of the total
variance of the variables in the model (Brown, 2001).

Further analysis of each factor was done to determine changes in individual
symptom scores and Quality of Life Index (QLI) scores in the presence of increasing
numbers of symptoms in the grouping. As can be seen from Table 12, individual symptom
scores increased as numbers of symptoms in the cluster increased. The greatest increase in
scores occurred with the addition of a third symptom. QLI scores declined with increasing
numbers of symptoms. Figure 6 illustrates the relationship between numbers of symptoms
and QLI scores. From the graph it appears that the greatest decrease in QLI scores occurred

with the addition of a second symptom.

81

 

Table12

Change in Individual Symptom Scores with Increasing Numbers of Symptoms in the

 

 

 

Cluster
Total symptom score
1 Symptom 2 Symptoms 3 Symptoms

Factor 1

Tiredness 13.33 13.93 16.15

Problems sleeping NA 15.29 16.35

Muscle weakness NA NA 17.15
Factor 2

Shortness of breath 11.03 14.77 NA

Chest pain NA 1 2. l 5 NA
Factor 3

Joint pain 14.76 15.13 17.00

Itching NA 13.73 16.00

Headache NA NA 14.29
Factor 4

Nausea/vomiting 1 1.29 1 1.92 NA

Abdominal pain NA 1 3 .69 NA

 

Note: NA = not applicable

82

 

30-

25“

20“

10-’

 

 

 

 

 

 

 

Factor 1 Factor 2 Factor 3

 

 

Factor 4

 

 

I 0 Symptoms
Ell Symptom

D 2 Symptoms
I 3 Symptoms

 

 

Figure 6. Quality of life scores relative to the number of symptoms in the cluster

reported.

83

 

Question 3: What strategies are used by ESRD hemodialysis patients to relieve
symptoms? Patients used a variety of both self-initiated and prescribed methods in their
efforts to obtain symptom relief. As can be seen in Table 13, medication and
rest/relaxation were the strategies most commonly employed to relieve an assortment of
symptoms. Medication was used most frequently to relieve different types of pain as well
as itching and problems sleeping. Often medication was used in combination with other
methods for more effective relief. For example, the treatment regimen for headache
frequently included both medication and rest/sleep/relaxation. The majority of patients
(55%) used rest/sleep/relaxation for relief of tiredness. Twenty-four percent of patients
who complained of difficulty sleeping used distraction as a method of coping. However,
this strategy was not effective in terms of symptom relief.

Patients used the widest array of strategies to relieve muscle cramps. Rather than
medication, they used a variety of alternative techniques including: stretching, walking,
and massage. Indicating the removal of too much fluid, cramps that occurred during
dialysis sessions were relieved by the administration of saline intravenously. The most
original method used by patients was eating pickles. Pickles were thought to replace
sodium and water lost with overly aggressive fluid removal during hemodialysis.

For all symptoms assessed, a portion of the patients did nothing to obtain relief.
This was especially noteworthy in the case of muscle weakness, the symptom that
received the highest symptom score. Of the patients who reported muscle weakness, 24%
indicated that they used no specific strategy to increase strength. Eighteen percent of the
patients stated that they exercised to build strength, but none did so with any regularity.

Refer to Table 13 for a summary of methods used to relieve all symptoms studied.

84

 

Table 13

Strateg’es Used by Patients to Relieve Symptoms

 

 

Symptom
n (valid %)*
Relief Strategy Tiredness Itching Headache Problems Joint Cramps
sleeping pain
Nothing 18 (18) 5 (9) 2(5) 14 (17) 11 (18) 5(7)
Medication 9 (9) 28 (50) 33 (85) 47 (57) 39 (62) 20 (29)
Rest/sleep/ 72 (72) 0 18 (46) 10 (12) 15 (24) 4 (6)
relaxation
Exercise 8 (8) NA NA NA 9 (14) 23 (34)
Distraction 13 (13) 0 0 31 (89) 1 (2) 0
Skin care NA 28 (50) NA NA NA NA
Cut out NA NA NA 3 (4) NA NA
daytime nap
Cut out caffeine
and alcohol NA NA NA 1 (1) NA NA
Change position NA NA NA NA NA NA
Change
breathing NA NA NA NA NA NA
pattern
Change eating NA NA NA NA NA NA
pattern
Other 8 (8) 20 (36) 9 (23) 35 (43) 18 (29) 46 (68)

 

Note. * n (valid %) = number and percent of patients reporting the symptom who used the

strategy. NA = not applicable to the symptom.

85

Table 13 (continued)

 

 

 

Symptom
n (valid %)*

. Shortness Nausea Abdominal Muscle
Relief Strategy of breath Chest pain vomiting pain weakness
Nothing 8 (19) 3(18) 14(11) 6 (33) 31 (46)
Medication 6 (14) 10 (59) 18 (44) 8 (44) 5 (8)
Rest/sleep/ 24 (56) 4 (24) 5 (12) 4 (22) 15 (22)
relaxation
Exercise NA NA NA NA 23 (34)
Distraction 1 (2) 0 2 (5) 0 0
Skin care NA NA NA NA NA
Cut out NA NA NA NA NA
daytime nap
Cut out caffeine
and alcohol NA NA NA NA NA
Change position 8 (19) 2 (12) NA 3 (17) NA
Change
breathing 4 (9) NA NA NA NA
pattern
Change eating NA NA 9 (22) NA NA
pattern
Other 13 (30) 3 (18) 6 (15) 4 (22) 3 (5)

 

Note. * n (valid %) = number and percent of patients reporting the symptom who use the

strategy. NA = not applicable to the symptom.

86

Question 4: What level of symptom relief is achieved by ESRD hemodialysis
patients? Refer back to Table 9 for the total relief scores for all symptoms assessed.
Possible scores ranged from 0 to 30. For this sample, relief scores ranged from 5.55 to
19.60 (M = 14.48, SD = 6.73), with higher scores indicative of greater relief. The
symptom with the highest symptom score, muscle weakness was also the least effectively
relieved (M = 5.55, SD = 8.40). The related symptoms of sleeping problems (M = 11.20,
SD = 10.35) and tiredness (M = 12.23, SD = 9.23) were ranked as the second and third
least effectively relieved. Although rated as the most severe, cramps received one of the
lowest symptom scores and the highest relief score.

To determine the extent to which patients were satisfied with the level of relief
they obtained from symptoms, the scores on the satisfaction subscale of the symptom
relief tool were examined. Scores for individual symptoms-ranged from 1.17 to 3.84,
rated on a 0-5 point scale with higher scores indicative of greater satisfaction. Patients
were the least satisfied with relief from muscle weakness, most satisfied with relief from
cramps. Mean satisfaction scores for the remaining symptoms included: sleeping
problems (2.27), nausea/vomiting (2.49), abdominal pain (2.61), tiredness (2.68), joint
Pain (2.92), chest pain (3.25), itching (3.29), shortness of breath (3.47), and headache
(3.80). Again, higher scores correspond to greater relief.

Question 5: What is the relationship between influencing factors and symptoms
and the QOL reported by ESRD hemodialysis patients? Correlation and regression
analysis were used to examine the relationship among influencing factors, symptoms, and
QOL. The correlations among these variables are presented in Table 14. Special note is

made of the correlations between influencing factors and QOL and symptoms and QOL.

87

:00 w a .1. .80 .v. :0 .. .0002

 

 

om:
-- mo 320:0
208 women
0:.0- -- 880800
280
3.00. 3.00.0 -- 8000800
0:206:00
RN? .0 a No 3.3.: -- “058800
300.0. 10.0 ......00.0 :0 -- 8:088:
1.00.0- 0:.0 $00.0 00.0 $000 -- 00280.
o a .0 mod- 0 H .o- :6. mod- mod -- 83:
mod 0. a .o 3 .o mod 0 a .o- m H .o mod -- 03050
from: o a .o- immo- mod *0 H .o- 1.9.:- immo- mod- -- ow<
em: :0 880 mono: 6.800 0:206:00 5:000:95 beta/00 comm 00:50 om<
320:0 Sofie-Sm EoEEzm 2:00:80

 

orb-NKQ bagm nee KEEN ESQ-Ekwko 38M 60:93:00? assoc-m $0.805:ng nmEEt-m M08500 0.200.033.0000

3 2an

88

tionship among influencing factors and symptoms has been previously discussed.
LS significantly and positively related to age. As age increased, QOL improved.
depression, comorbid conditions, and symptom scores were all significantly and
[y correlated with QOL. Increasing anxiety, depression, numbers of comorbid
rs, and symptom scores were accompanied by a decrease in self—reported QOL.
eparate regression analyses were conducted to determine which symptom
ns and symptom clusters (factors 1-4 identified earlier) were significant
5 of QLI scores. QLI scores were regressed on the mean scores for each of the
dimensions. Of severity, frequency, duration, and distress, only distress was
significantly predict QLI scores ([3 = -0.36, t(124) = -3.08, p = 0.00). Greater
distress was associated with diminished QOL. Twenty percent of the variance
ores was accounted for by the model; distress accounted for 6%. In addition, it
1 that tolerances ranged from 0.48-0.67, providing evidence of shared variance
3 dimensions.
91 scores were also regressed on the four factors (clusters) identified with the
f symptom concurrence. Prior to performing the regression analysis, a score
uted for each factor. The factor score was simply the average of the symptom
iprising the factor. Factor 1, including tiredness, sleeping problems, and
alcness, was the strongest predictor of QLI scores ([3 = -0.40, t(125) = -4.85,
The presence of this cluster of symptoms was related to lower QOL.
rltiple regression was again used to examine the impact of influencing factors
)ms on QOL. The results of this analysis with QOL as the dependent variable

ed in Table 15. QOL was first regressed on the situational/demographic,

89

Table 15

Summary of Regression Equations Identzfi/ing Predictors of Quality of Life

 

 

Regression Independent R2 B SE B t p
variables

1 Influencing 0.67
factors
Age 0.09 0.02 0.27 4.46 0.000
Gender 0.24 0.55 0.02 0.44 0.663
Race 1.61 0.62 0.15 2.61 0.010
Anxiety -0.22 0.07 -0.21 -2.96 0.004
Depression -0.69 0.08 -8.40 -8.40 0.000
Comorbid -0.26 0.14 -0. 10 -1.90 0.060
conditions

2 Influencing
factors, symptom 0.67
scores
Age 0.09 0.02 0.27 4.41 0.000
Race 1.60 0.62 0.15 2.57 0.011
Anxiety -0.21 0.08 -0. l 9 -2.65 0.009
Depression -0.70 0.08 -0.55 -8.65 0.000
Comorbid -0.25 0.15 -0. 10 -1.69 0.094
conditions
Symptom scores -0.00 0.01 -0.01 -0.13 0.896

3 Influencing
factors, symptom
scores, level of 0.69 _
relief .
Age 0.09 0.02 0.28 4.66 0.000
Race 1.73 0.62 0.16 2.81 0.006
Anxiety -0. l 7 0.08 -0. l 6 -2.23 0.028
Depression -0.70 0.08 -0.56 -8.86 0.000
Comorbid -0.28 0.15 -0.11 -1 .91 0.058
conditions
Symptom scores -0.01 0.01 -0.09 -1.08 0.284
Level of relief 0.01 0.01 0.10 1.63 0.105

 

90

Q

Table 15 (continued)

 

 

Regression Independent
variables R2 B SE 13 t p
4 Influencing
factors, symptom
scores, level of
relief, interaction 0.68
Age 0.10 0.02 0.32 5.35 0.000
Race 1.65 0.63 0.15 2.64 0.009
Depression -0.78 0.08 -0.62 -10.30 0.000
Comorbid -0.26 0.15 -0. 10 -l .83 0.070
conditions
Symptom scores -0.02 0.01 -0. 16 -2.24 0.027
Level of relief 0.03 0.01 0.20 2.92 0.004
Interaction -0.0004 0.00 -0. 12 -2.1 1 0.037
5 Influencing

factors, symptom
scores, level of
relief 0.66
Age 0.10 0.02 0.31 5.13 0.000
Race 1.45 0.63 0.13 2.31 0.023
Depression -0.76 0.08 -0.60 -9.96 0.000
Comorbid —0.25 0.15 -0.10 -1.69 0.094
conditions
Symptom scores -0.03 0.01 -0.17 -2.34 0.021
Level of relief 0.02 0.01 0.14 2.25 0.027

 

91

psychological, and physiological influencing factors that were previously identified as
predictors of symptoms. The model that included influencing factors resulted in

R2 = 0.67, F(6,122) = 41.89, MSE = 8.82, p = 0.00. Age, race, anxiety, and depression
were significant predictors of QOL. Older age and Black race predicted higher levels of
QOL. Conversely, higher levels of anxiety and depression predicted lower levels of
QOL. Gender was removed from the model at this point because it was nonsignificant at
p 3 0.50.

Symptom scores were then added to the regression. See Table 15, regression 2.
This model accounted for 67% of the variance in QOL. Symptoms did not predict any
unique variance in QOL over and above the influencing factors (R2 = 0.67,

F(6,122) = 41.80, MSE = 8.83, p = 0.00). Age, race, anxiety, and depression were again
significant predictors of QOL. Symptom scores alone had no impact. However, symptom
scores were retained in the model to examine their influence with the addition of level of
relief.

Question 6: What is the moderating effect of level of symptom relief on the
relationship between symptom scores and QOL? Level of relief was added to the
regression model as a first step in exploring its potential moderating effect on the
relationship between symptoms and QOL. As with symptom scores, a total relief score
was computed for all participants by summing relief scores for all reported symptoms. As
can be seen in regression 3 on Table 15, neither symptom scores nor relief scores were
significant predictors of QOL. The addition of level of relief increased the amount of
variance accounted for in QOL by 2%. However, symptom scores moved closer to

significance. Anxiety was removed from the model following this analysis because of

92

 

10W tolerance (0.49). Low tolerance indicated that much of the variance was shared by

01 . . . . . .
bervanables. As noted in Table 14, anxrety was highly correlated wrth depressron,

symptom scores, and QOL.
The interaction between symptom scores and level of relief was then included in

the model to determine if level of symptom relief had a moderating influence on the
relationship between symptom scores and QOL (see Table 15, regression 4). In this

model, age, race, depression, symptom scores, and level of relief were significant
predictors of QLI scores. The interaction between symptom scores and level of relief was
significant as well. However, the interaction added approximately only 1% of the

variance to the model. The meaning of the interaction was also not clear and
counterintuitive conceptually. The interaction indicated that low symptom scores were
related to low QLI scores when level of relief was high. In an attempt to better
understand the interaction, QLI scores were regressed on symptom scores, level of relief,
and the interaction. Both symptom scores and level of relief remained significant

predictors of QOL but the interaction was nonsignificant (B = 0.00, t(125) = 0.01,

p = 0.995). Because of this and the percent of variance it accounted for in QLI scores, the
interaction was removed from the final model.

In the final model, age, race, depression, symptom scores, and level of relief were

significant predictors of QOL, accounting for 66% of the variance in QOL scores. It was

found that higher QOL was predicted by older age, non-White race, and lower levels of
depression and physical symptoms. The standardized coefficient for age was [3 = 0.31,

t(122) = 5.13, p < 0.00. A person who was one standard deviation above the mean on age

was 0. 31 standard deviations above the mean on QOL. Race also significantly predicted

93

 

QOL, with non-Caucasian patients reporting higher levels. Both depression (B = -0.60,
.022) = -9.96,p < 0.00) and symptom scores ([3 = -0.l7, t(122) = -2.34,p = 0.02)
negatively influenced QOL. Individuals who were one standard deviation above the mean
m depression and symptom scores were 0.60 and 0.17 standard deviations below the
nean on QOL respectively.

The main effect of level of relief on symptoms was positive, such that higher
evels of relief predicted higher QOL. Patients who were one standard deviation above
he mean on level of relief were 0.14 standard deviations above the mean on QOL. In
ight of the results presented, the original theoretical model underlying this study required

'evision. See Figure 7 for the revised model.

94

 

 

 

.400 one 088mg? 52:0 832:8: 058$:me .0... .082

3:08 303885 come/om .b ocswE

 

000200 02:2 as,
women |.V eonoflmnam

/ \

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

momwofibm nomad
”HE-EM SCH-Em m0 Amt/m4

80m .

ow< ..

oEmfiwoaoeaaouaBmm

map—beam 02575052

30860839 -

A 882 .

Enema—3900A

has”: . 8:258:00 Seam ._

unconcem one 880m .. 002005 Sofia—Sm .. mMOHU<m 67:0qu?
_mBEmmhaomwBonosmm . L song—D
manouogm :80 0233 - >826on
00000 00:80 00850

880.580 Eocam .
mac
8725088 2.008 038800
mEO-wgm 5:250 8805 .
Fame—203m
mafia-000m 0200qu2

 

 

 

95

Chapter 5
DISCUSSION

The potential need for palliative care in the end-stage renal disease (ESRD)
population has gone unrecognized by the general healthcare community as well as by
those who reimburse for care. End-stage renal disease is a life—limiting illness that places
tremendous physical and psychosocial burdens on both patients and families. Despite this
fact, hemodialysis patients rarely benefit from services that focus on the unique needs of
individuals who anticipate a shortened life span. The inability to access palliative care
services occurs for a number of reasons including continued active treatment with
hemodialysis, lack of financing, and absence of documentation of the benefits of
palliative care to ESRD patients (Moss & ESRD Workgroup, 2001). In addition, there is
insufficient empirical evidence to confirm the need for palliative care, especially with
respect to symptom relief. Such data are needed to spotlight the issue and capture the
attention of healthcare providers, government officials, and private insurers who make
crucial decisions regarding reimbursement of services.

This investigation set out to validate the need for palliative care in the ESRD
hemodialysis population, specifically in the area of symptom relief. To this end,
documentation of the total symptom experience associated with ESRD became the
central purpose of the study. Factors that influence symptoms as well as their
consequences in terms of QOL were investigated. Extending the examination of symptom

experience a step further than in previous research, the degree to which patients obtain

relief from their unpleasant symptoms was also assessed.

96

Influencing factors

The exploration of the symptom experience of hemodialysis patients began with
an analysis of a number of factors thought to influence symptoms. Knowledge of the
physiological, psychological, and situational/demographic variables that are linked to
symptoms enhances the ability of healthcare providers to anticipate, prevent, and relieve
them. While some of these factors are amenable to intervention, others are not. However,
it is important to consider those factors that cannot be altered because they help to
identify patients who are at risk for higher levels of symptoms.

The physiological variables of interest in this investigation were disease severity,
defined by degree of metabolic homeostasis and comorbid conditions, and adequacy of
dialysis treatments. Low Chemistry Abnormality Scores and Kt/V values within the
desired range (i.e., values 3 1.2) suggest that hemodialysis was effectively maintaining
homeostatic balance for the majority of patients in this sample. Therefore, it was not
surprising that these summary indicators of disease severity and dialysis adequacy were
not significant predictors of high symptom scores. This finding is noteworthy in that
healthcare providers may presume that patients whose laboratory values are within an
acceptable range do not experience high levels of symptoms and may result in failure to
identify those at risk.

To further investigate the relationship between laboratory values and symptoms,
individual blood tests were examined more closely. Only phosphorus and alkaline
phosphatase levels exceeded the acceptable range for dialysis patients and neither was
significantly related to total symptom score (Good Samaritan Hospital Chronic [Dialysis

Center, 2003). Phosphorus was related to itching, a symptom commonly thought to

97

accompany elevated levels (Daugirdas & Ing, 1994; Vanholder, Glorieux, De Smet, &
De Deyn, 2004). However, J akic ( 1999) and Bruni et a1. (1989) reported no relation
between this laboratory parameter and itching in the hemodialysis patients they studied. It
has been suggested that joint pain is associated with increased levels of alkaline
phosphatase (Colorado HealthSite, 2004). Although Barrett et al. (1990) validated this
relationship, the link between elevated alkaline phosphatase levels and joint pain was not
supported in this study. In light of conflicting results of this and previous research, the
impact of biochemical abnormalities on the symptoms experienced by hemodialysis
patients remains in question.

Of the physiological influencing factors examined, only number of comorbid
conditions significantly predicted symptom scores. This is a particularly relevant finding
in that ESRD patients experience a high level of morbidity despite technically adequate
dialysis treatment (Poppel et al., 2001). According to the United States Renal Data
System (2001), 50% of incident patients have three or more comorbid conditions when
they begin therapy. As time on dialysis progresses, increasing numbers of concurrent
illnesses emerge that compound symptom burden. In this sample of prevalent patients,
85% eXperienced between 3 and 10 additional comorbidities. Cardiovascular disorders,
pefipheral vascular disease, diabetes, chronic lung disease, and arthritis were among the
conditions that patients experienced, each adding its own constellation of symptoms to

the total burden.
The influence of comorbid conditions on symptoms reported by patients in this
study is difficult to corroborate with previous research. Barrett et a1. (1990) failed. to

demonstrate a positive relationship between the two factors in a study of hemodialysis

98

patients. On the other hand, Merkus et al. (1999) found that increasing numbers of
comorbid illnesses in combination with advanced age lead to higher symptom burden,
accounting for 12% of the variance in total symptom score. Although age was not a
significant predictor of symptoms in this study, the positive influence of comorbid
conditions on symptoms was supported. An additional 7% of the variance in symptom
score was explained by the inclusion of comorbid conditions in the model. Although the
positive relationship between comorbid conditions and symptoms seems intuitive,
conflicting results and limited research make it impossible to conclude with certainty that
a connection consistently exists.

Establishing a relationship between psychological influencing factors and
symptoms is complicated as well, confounded by the overlap of physical symptoms of
illness and psychological distress. However, the link between psychological status and
symptoms has been established with greater reliability through the use of measures that
eliminate items that are symptomatic of both physical and mental disorders.

The Hospital Anxiety and Depression Scale (HADS) is an example of a tool that
relies only on psychic symptoms to assess psychological distress (Zigmond & Snaith,
1983). Because dialysis patients suffer from a number of physical symptoms that are also
indicative of psychological disorders, researchers have used this instrument to assess
anxiety and depression in this population. In all cases, anxiety and depression were
significantly correlated to physical symptom scores (Killingworth & Van Den Akker,
1996; McCann & Boore, 2000).

Using the HADS to measure anxiety and depression, the findings of this

study fiirther validate the connection between psychological status and physical

99

symptoms. Significant correlations were found between anxiety and depression and both
total symptom scores and nearly all of the individual symptom scores. In addition,
anxiety and depression together were the strongest predictors of symptoms, accounting
for 24% of the variance in symptom scores. These findings provide evidence of the
dynamic nature of the mind-body interaction. The data also lend insight into areas where
interventions by healthcare providers may improve the comfort and well-being of
patients.

The unanswered question of cause and effect complicates the choice of the most
appropriate intervention. The theoretical model for this study depicts a linear relationship
between psychological factors and symptoms, with the implication that anxiety and
depression lead to higher symptom scores. However, the cross-sectional nature of this
study does not allow for determination of a causal relationship between psychological
factors and somatic symptoms. This is due, in part, to the inability to determine the
temporal ordering of variables with a cross-sectional design.

Furthermore, recent research with other patient populations does not uphold the
inference of a causal link from psychological status to physical symptoms. In a recent
study, structural equation modeling was used to test the Theory of Unpleasant Symptoms
in a sample of newly diagnosed lung cancer patients (J ablonski, Gift, von Eye, Given, &
Given, 2004). Rather than depression directly affecting symptoms, the opposing
relationship was noted. Higher levels of symptoms resulted in less positive mental health
and higher levels of depression. The direction of this relationship was supported in a
longitudinal study conducted by Kurtz and colleagues (2002), who found that symptom

severity was a significant predictor of depressive symptoms in elderly lung cancer

100

patients. In light of these findings, longitudinal rather than cross-sectional research is
needed to clarify the direction of the relationship between psychological status and
symptoms in the ESRD hemodialysis population.

Finally, none of the situational/demographic influencing factors (age, gender,
race, marital status, level of education, length of time on dialysis) included in this study
were significantly related to symptom scores. McCann and Boore (2000) assessed the
impact of these same factors on the fatigue experienced by hemodialysis patients with the
identical results. Only age was positively related to nausea and headaches in a study
conducted by Barrett et al. (1990). Merkus and colleagues (1999) were unable to
demonstrate an independent effect of demographic characteristics on level of
symptomatology as well.

These results question the need to include this category of influencing factors in
the theoretical model underlying this study. However, it may be that situational variables
that do affect symptoms have not yet been identified. Lifestyle factors such as diet and
exercise, availability and access to healthcare resources, and social support are examples
of situational factors that conceivably impact symptoms but were not examined in this
investigation (Kimmel et al., 1995). Exploration of the effect of a greater variety of
situational/demographic factors is needed prior to removing this category of influencing
factors from the model.

Symptoms

Crucial to any effort to validate the need for palliative care services in the ESRD

population is an in-depth analysis of the symptoms experienced by patients. To this end,

the number, variety, and nature of symptoms reported by patients in this sample were

101

systematically explored. Data indicated the presence of a relatively high number and
variety of symptoms. Of the 11 physical symptoms routinely assessed, patients recounted
an average of five. Upon further questioning, patients also reported a wide assortment of
additional symptoms involving several body systems. Among them were changes in taste,
smell, and vision (sensory); anorexia, thirst, and constipation (gastrointestinal); and
numbness/burning in hands/feet and restless legs (peripheral vascular and neurological).

As anticipated, the prevalence of symptoms in this sample mirrored that reported
in the literature. Tiredness, problems sleeping, cramps, muscle weakness, and numbness
in the hands/ feet were the top five symptoms experienced by patients in this study.
Although rankings varied slightly, the majority of studies also reported these symptoms
among the most commonly occurring (Barrett et al., 1990; Curtin, Bultman, & Thomas-
Hawkins, 2002; Killingworth & Van Den Akker, 1996; McCann & Boore, 2000; Merkus
et al., 1999; Parfrey et al., 1988). Nausea/vomiting, chest pain, abdominal pain, and
headaches, while less prevalent, were still reported by a fair number of patients in this
and previous studies (Merkus et al., 1999; Parfrey et al., 1988). The presence of an array
of additional symptoms such as those noted in this study has been documented in the
literature as well, providing further evidence that multiple symptoms are a common
occurrence in the lives of hemodialysis patients (Curtin, Bultrnan, & Thomas—Hawkins,
2002)

A thorough investigation of the symptoms experienced by hemodialysis patients,
however, requires more than knowledge of their prevalence. In this study, analysis of
symptoms extended beyond prevalence to an examination of their characteristics as well.

Previous research has typically focused on limited aspects of symptoms, primarily their

102

 

occurrence and severity or frequency (Curtin et al., 2002; Killingworth & Van Den
Akker, 1996; Parfrey et al., 1987). In this study, a multidimensional profile was
constructed for each of the 11 physical symptoms including its severity, frequency of
occurrence, duration, and associated distress. A total score was also calculated for each
symptom by summing the ratings given on each of these four dimensions.

An advantage of this summary score was that it provided a means by which to
estimate how problematic each symptom was both individually and relative to other
symptoms. Individual symptom scores revealed that patients in this sample experienced
high levels of symptoms. All symptoms were rated above the midpoint of the possible
range of scores, 0 to 20. Muscle weakness received the highest score at 16.26. Not
surprising, tiredness and problems sleeping also received high scores. Nausea/vomiting
received the lowest score at 11.49. Having access to summary symptom scores is
beneficial in that it allows healthcare providers to estimate the relative importance of
each symptom to the patient and prioritize interventions to relieve them.

A disadvantage of the summary scores, however, was that they obscured the
unique contribution of each of the symptom dimensions. For example, in this study
nausea/vomiting and chest pain received similar mean symptom scores, 11.49 and 11.53
respectively. However, ratings given the individual dimensions were dissimilar. Although
chest pain occurred infrequently and was of moderate severity, it lasted as much as half
the day and was perceived as severely bothersome. Nausea/vomiting, on the other hand,
occurred more often and with greater severity, but it did not last as long and was only
moderately bothersome. Summary symptom scores permitted a judgment as to how

problematic these symptoms were in general terms but the relative influence of each

103

dimension could not be established.

Severity, frequency, duration, and distress ratings were examined for each of the
symptoms in an effort to determine the differential impact of each dimension on the
experience of the symptoms. However, it was not possible to distinguish any pattern
among the individual dimension ratings. No one dimension was consistently rated higher
or lower than the others and three of the four dimensions were significantly
intercorrelated. These findings suggest that the experience of a symptom is shaped by the
interplay of all of its dimensions, making it difficult to disentangle the influence of one
dimension from the other. This conclusion draws attention to the need for
multidimensional symptom assessment.

Symptom assessment is typically limited to severity, a practice that is reinforced
in the clinical setting. A prime example is the widespread measurement of pain using a
10-point severity scale. This routine not only ignores the patient’s reactions to how often
pain occurs and how long it lasts, but also discounts the emotional response to the
- symptom. Assessment of both the cognitive (i.e., perception of severity, frequency and
duration) and emotional (i.e., perception of distress) response to symptoms is
fimdamental to symptom management efforts. Multidimensional assessment contributes
data that aids the design of interventions that treat the symptom from multiple angles,
hopefully resulting in greater relief. Despite the potential advantages, measurement of all
symptom dimensions is not yet standard clinical practice.

To fiirther explore the benefits of multidimensional symptom assessment,
symptom dimensions were also examined in relation to QOL, the outcome of interest in

this study and a major focus of palliative care. Only the degree of distress associated with

104

 

 

 

symptoms was found to be a significant predictor of QOL, accounting for 6% of the
variance in Quality of Life Index scores. This relationship between symptom distress and
QOL was recognized by McDaniel and Rhodes (1995) in an earlier analysis of symptom
experience. In their middle-range TOUS, Lenz and colleagues (1997) also suggested that
it is symptom distress that most contributes to QOL. The results of this study lend
strength to these theoretical claims.

The finding that symptom distress predicts QOL also reinforces the importance of
this dimension as a focus of assessment and intervention. In addition, McDaniel and
Rhodes (1995) emphasized the importance of assessing symptom distress to identify the
patient’s affective response to symptoms, to motivate self-management of symptoms, and
to evaluate the effectiveness of different modes of treatment. However, especially in
hemodialysis clinics where one nurse is responsible for several patients, the technical
aspects of care critical to maintain patients’ lives often overwhelm concern for the QOL
they experience. Frequently the time needed to thoroughly assess all symptom
dimensions as well as plan and implement appropriate interventions is not available.
Although not evaluated in this study, it is also conceivable that the staff do not possess
the expert knowledge necessary to provide this care. This reality adds leverage to the plea
for access to and reimbursement for palliative care services in this population.

Finally, the analysis of symptoms reported by patients in this study was
completed with an examination of the dimension of concurrence. Interest in this
dimension has gained momentum in recent years, as evidenced by increasing numbers of
conference presentations and papers published on the topic (Dodd et al., 2001b; Gifi,

Stommel, J ablonski, & Given, 2003; Miller, Nail, Rosenfeld, & Perrin, 2004). Research

105

investigating symptom concurrence (i.e., clusters), however, is still in its infancy despite
this emergent attention. Furthermore, this study represents the first attempt to explore this
dimension in the ESRD population.

According to Dodd et al. (2001b), a symptom cluster is defined as three or more
concurrent symptoms that are related to each other. In addition, convention indicates that
a factor (i.e., cluster) should be composed of a minimum of three variables (Factor
analysis, 1995). In this study, four groupings emerged from the 11 physical symptoms
assessed including: 1) energy/vitality, 2) cardiac functioning, 3) pain/comfort, and 4)
gastrointestional functioning. Not all of these groups of symptoms qualified as clusters as
defined by the criteria noted previously. It appeared that the limited number of symptoms
included in the analysis was responsible for two factors being defined by only two
symptoms. However, since the symptoms in each factor were theoretically related and
demonstrated acceptable factor loadings, they were considered clusters in this study.

It has been hypothesized that concurrent symptoms not only interact to alter the
perception of each individual symptom but also impact outcomes in a synergistic manner
(Dodd et al., 2001b; Lenz et al., 1997). The clusters identified in this study supported this
hypothesis to a degree. Individual symptom scores increased as numbers of symptoms in
the cluster increased. In contrast, QOL decreased in the presence of each additional
symptom in the cluster. All of the clusters followed this pattern.

This relationship has been documented in previous studies, but with cancer
patients. In a study of newly diagnosed elderly lung cancer patients, Gift and colleagues
(2004) reported a strong positive correlation between number of symptoms in the cluster

and severity of symptoms. An association between increasing numbers of symptoms and

106

patient outcomes was also noted. Patients who reported higher numbers of symptoms in a
given cluster also experienced decreased physical functioning. Similar findings have also
been reported by Dodd et al. (2001b), noting diminished functional status in the presence
of the cluster including pain, sleep insufficiency, and fatigue.

Also of interest was the link between each of the clusters and QOL. In this
analysis, rather than examining the impact of individual symptoms on the cluster, the
influence of the cluster as a unit was assessed. The energy/vitality cluster, including
tiredness, sleeping problems, and muscle weakness, was the strongest predictor of QOL.
Given the number of patients reporting these symptoms as highly problematic, this
finding has significant clinical ramifications. It is suggested that interventions targeted
toward this cluster of symptoms may yield a dramatic improvement in the QOL of
hemodialysis patients. However, research is needed to test this hypothesis.

Based on the results of this study, it appears as though hemodialysis patients
'suffer from a substantial number and wide variety of unpleasant symptoms. In addition,
symptoms that co-occur further increase symptom burden and diminish QOL. The
distress associated with the symptoms adversely influences QOL as well. In light of these
findings, a key end result of this exploration of symptoms is a beginning validation of the
need for palliative care in the hemodialysis population.

Level of Symptom Relief

Knowledge of the prevalence and characteristics of symptoms that accompany
ESRD is insufficient to build a definitive case for palliative care. The extent to which
symptoms are relieved is of paramount importance to the argument as well. A deficiency

in existing theoretical models of symptom experience and symptom management is that

107

 

 

this concept is not included as a key component. Neither the TOUS (Lenz et al., 1997)
nor the Symptom Management Model developed by Dodd and colleagues (2001a)
explicitly incorporates level of relief as an outcome. As a consequence, the current
research literature lacks documentation of the level of relief
patients obtain from their symptoms. Since this information is necessary to validate or
refirte the need for palliative care, level of relief was a focus of this investigation.
Hemodialysis patients in this study experienced only fair relief from their
symptoms. This conclusion is supported by a mean relief score of 14.48, which is well
below the maximum possible of 30. Further analysis of individual symptoms suggested
that the vaguest symptoms in terms of etiology and treatment were also the least well
relieved. Muscle weakness, problems sleeping, and tiredness are examples of symptoms
that have multiple causes but few effective remedies. These symptoms ranked the highest
in terms of symptom scores and the lowest with respect to relief. Interestingly, they also
formed the cluster of symptoms that had the greatest impact on QOL. These findings
point to an urgent need to target this group of symptoms for intervention and research.
Upon examination of the satisfaction component of the relief scores, it became
apparent that patients were not entirely satisfied with the amount of relief they obtained
from their symptoms. However, anecdotal comments made by patients during the
interviews raised doubt as to validity of the satisfaction scores they reported. Their
remarks suggested that resignation or acceptance better described their response to level
of symptom relief than satisfaction. When questioned regarding satisfaction with relief,
one common response was “I guess I’m satisfied, what else can I do about it.” This

comment implied that patients were resigned to live with symptoms they believed to be

108

untreatable. An equally frequent comment, “You just take it as it is because it’s all a part
of the illness,” alluded to acceptance of symptoms that were thought to be unavoidable
consequences of ESRD and its treatment. In light of these statements, patients’
satisfaction scores may, in fact, reflect level of resignation or acceptance rather than
satisfaction.

This interpretation suggests one explanation for why some of the patients in this
study did nothing to manage their symptoms. For example, 24% of the patients who
complained of muscle weakness employed no particular tactic to improve their strength.
If the previous assumption is correct, the question arises as to what factors lead to
feelings of passive resignation and unquestioning acceptance of symptoms. While some
of the patients in this study used a variety of strategies to relieve symptoms, others did
not. The physiological, psychological, and situational/demographic factors that contribute
to this discrepancy in behavior among patients were beyond the scope of this study and
are yet to be determined.

Documentation of the need for palliative care in the hemodialysis population was
fiirther advanced by this examination of the level of symptom relief reported by the
patients in this study. To this point in the discussion, it has been determined that patients
experienced a variety of symptoms, several of which were not effectively relieved.
Furthermore, patients were not completely satisfied with the level of relief they obtained
from their symptoms. The argument for palliative care is not complete without
examination of the impact of symptoms on QOL.

Quality of Life

The ultimate outcome of interest to palliative care is the QOL of patients who live

109

 

in the shadow of a life-limiting illness. Symptoms are a major focus of concern since they
have been shown to have a significant negative effect on QOL. Previous research has
documented the impact of symptoms on the QOL of patients with ESRD. Without
exception, these studies have demonstrated that distressing symptoms detract from
dialysis patients’ well-being and enjoyment of life (Curtin et al., 2002; Killingworth &
Van Den Akker, 1996; Merkus et al., 1999; Parfrey et al., 1988). It was not surprising,
then, when the results of this study reinforced this relationship between symptoms and
QOL. Patients with high symptom scores reported lower QOL than those with low
symptom scores.

The QOL of patients in this study was also influenced by how well their
symptoms were relieved, a relationship that has not been previously reported in the
literature. Symptom relief scores were significant predictors of QOL scores such that low
levels of symptom relief were predictive of diminished QOL. This finding causes one to
View symptom experience from a slightly different perspective than in the past. The
primary focus of attention becomes level of relief, the direct anticipated outcome of
intervention, and its indirect beneficiary, QOL. This does not detract from the importance

of the relationship between symptoms and QOL, but spotlights the actions required to

bring about relief.

Limitations

While this study contributes to an improved understanding of the symptom
experience and QOL of hemodialysis patients, findings must be considered in light of its
limitations. A number of design and methodological issues are acknowledged for this

reason.

110

The cross-sectional design of this investigation prevents the confirmation of
causal links between variables. A significant correlation between variables suggests only
that they are related and the direction of the relationship. It cannot be stated with any
certainty, for example, that high symptom scores caused diminished QOL even though
they were negatively related. Only longitudinal studies can demonstrate the temporal
ordering of variables necessary for validation of cause and effect.

Cautious interpretation of study results is also suggested due to the timing of data
collection. All participant interviews were conducted on treatment days during dialysis
sessions. Although consistent with previous research with hemodialysis patients, this
design may introduce a bias toward reporting symptoms that occur at one point in time.
Fluctuations in symptoms that may occur on off-dialysis days may not be captured as a
consequence. This hypothesis was supported in one study in which patients stated that
they feel so different during dialysis treatments that they doubt that responses made then
accurately reflect the interdialytic period (Meyer et al., 1994, p. 271). In this study,
patients were asked to report symptoms in relation to the past month to moderate the
impact of this variation in status. However, recall for past events has been shown to be
inaccurate and biased (Stone, Shiffrnan, Schwartz, Broderick, & Hufford, 2002). There
may also be a tendency for patients to focus on symptoms that are present at the moment
because they overshadow experiences of the recent past.

An additional limitation of this study related to the timing of data collection is the
use of laboratory values to compute Chemistry Abnormality Scores (CAS) that did not
coincide exactly with assessment of symptoms. Because blood is drawn routinely only

one time per month, very few of the patient interviews were conducted on the same day.

111

 

It is conceivable that blood chemistry values fluctuate from day to day. This variation in
metabolic status was validated to an extent in this study. There was essentially no
correlation found between the previous month’s and the current month’s CAS values

(r = -0.01, p = 0.95). As a consequence, the CAS computed from the most current
laboratory tests may not accurately reflect physiological status at the time of the
interview. This could potentially result in an inaccurate assessment of the impact of
abnormal blood values on symptoms. However, it was not feasible to interview all
patients on blood draw days.

The instruments used to measure symptoms and level of symptom relief must also
be mentioned as a study limitation. Both of the tools were developed specifically for this
study because the research questions could not be satisfactorily answered with existing
measures. However, both instruments performed well and produced the expected results.
Patients had no difficulty understanding or responding to individual items. Neither
measure took an inordinate amount of time to complete. However, it is suggested that
items evaluating relief relative to each symptom dimension could be removed without
loss of clinically relevant information. The overall relief item and the satisfaction item
resulted in sufficient data to determine level of symptom relief. Test-retest reliability for
each of the tools was adequate. Scores obtained for both symptoms and level of relief at
Time 1 were not significantly different from those at Time 2. A moderate correlation

between the two relief scores, however, suggests the need to further refine that

instrument.
Implications for Nursing Practice and Research

Despite its limitations, this study has implications for both clinical practice and

112

 

research relative to each component of symptom experience model that guides this study.
While some findings have immediate clinical applicability, others raise more questions
that require further research.

Influencing factors. Of all of the physiological, psychological, and
situational/demographic influencing factors examined in this study, anxiety and
depression were most strongly correlated to symptoms and QOL. Both disorders were
linked to increased levels of symptoms. The presence of depression also predicted
diminished QOL. Despite their negative impact on patients’ lives, routine screening for
anxiety and depression in hemodialysis patients is not typical of current clinical practice.
The reasons for this lack of attention to the psychological status of patients were not the
focus of this investigation. However, it is suggested that in the practice setting,
inadequate staffing levels force nurses to focus on the technical aspects of care to the
neglect of less urgent matters. Undocumented observations made in two busy dialysis
clinics during data collection suggest this to be the case. Nurses were justifiably
preoccupied supervising the technical aspects of dialysis for 6-12 patients, leaving little
time to handle less pressing issues. The initiation of a palliative care program in the
dialysis setting with staff who have time to attend to the psychological needs of patients
is pr0posed as one solution to this situation. There is an urgent need to take action to
improve patient outcomes.

While the findings of this study regarding anxiety and depression have direct
clinical application, research is needed to further clarify the impact of both physiological
and situational influencing factors on symptoms and QOL. Research to identify

additional physiological correlates of symptoms is recommended since only comorbidity

113

 

was shown to be significantly related in this study. For example, the influence
interdialytic weight gain, camitine deficiency, and medication regimens are examples of
variables not studied in this investigation that potentially alter symptoms and QOL.
Situational factors are an especially fertile area for research since many are amenable to
nursing intervention. As mentioned earlier, the influence of variables such as social
support, activity patterns, dietary intake, and type of artificial kidney used for dialysis
need to be explored. Of the situational variables examined in this study, none provided
direction for intervention to improve patient status. Both categories of influencing factors
require further exploration.

Symptoms. Some of the insights that emerged from this study relative to
symptoms can be confidently used in clinical practice. Other findings raise further
questions to be answered before application to patient care is warranted. The need for
multidimensional symptom assessment has been described. However, time constraints in
the clinical setting as well as the lack of quick, easy assessment tools limit the feasibility
of multidimensional evaluation at this time. Nonetheless, the addition of distress to
symptom assessment is recommended. The distress dimension was the only one that was
found to be a significant predictor of QOL. Including patients’ evaluations of the distress
associated with symptoms is straightforward and has the potential to provide valuable
information to guide interventions and enhance QOL.

Already alluded to, the availability of multidimensional assessment tools would
enhance nurses’ ability to completely evaluate patients’ symptoms. Research is needed to
develop these tools for use in both the clinical setting and in research endeavors. Since

multidimensional symptom assessment is new to most nurses, exploration of the most

114

effective methods for educating them as to its importance as well as its incorporation into
clinical practice is also essential.

Concurrence (i.e., symptom cluster) is a new area of research so it is not
surprising that it is a novel concept to practicing nurses as well. Beginning evidence
provided by this and a small number of previous studies suggests that it has relevance for
clinical practice. At this point, direct clinical application is limited. However, two
practical conclusions emerge from this investigation. First, nurses need to be aware that
the presence of co-existing symptoms may lead to a more intense symptom experience
that negatively impacts QOL. Secondly, one cluster of symptoms that emerged during
data analysis requires immediate attention in practice. Muscle weakness, tiredness, and
problems sleeping formed a cluster that was clearly troublesome to patients and was
negatively related to their QOL. The severity, frequency, duration, and distress associated
with these symptoms were ranked among the highest of the 11 physical symptoms
assessed.

Additional empirical exploration of the concept of concurrence is warranted to
further delineate its significance for clinical practice. This includes examination of the
consistency with which symptom clusters occur in multiple samples of hemodialysis
patients, the interaction of symptoms in the cluster, and the impact of symptom clusters
on patient outcomes. Refinement and standardization of the statistical methods used to
analyze symptom concurrence is also needed to enhance comparison of findings across
studies and patient populations.

Level of symptom relief. The symptom cluster comprised of muscle weakness,

tiredness, and problems sleeping was not only identified as the most troublesome in this

115

study but the least effectively relieved as well. It was also the cluster that was the
strongest predictor of diminished QOL. Therefore, it is imperative that nurses who have
frequent contact with hemodialysis patients focus on the management of these pervasive
symptoms. Doing so may significantly enhance the QOL of afflicted patients.

However, all three of the symptoms in this cluster are complex in terms of both
their etiology and treatment. Effective therapies are elusive. Clinical intervention studies
are needed to determine treatment modalities that will alleviate or, at a minimum,
diminish the impact of these symptoms. Although patients failed to obtain adequate relief
from other symptoms as well, muscle weakness, tiredness, and problems sleeping seemed
to be the most distressing to patients in this study.

Nurses are educated to provide holistic care. In the real world of everyday clinical
practice, however, insufficient time often limits their ability to do so. High nurse-patient
ratios and high illness acuity frequently result in greater emphasis on accomplishing tasks
than on meeting less immediate patient needs. The results of this study challenge nurses
to redirect their time and effort to two areas of particular importance to QOL,
psychological status and symptom management. Greater attention to the emotional and
physical comfort of patients may enhance their QOL. This investigation has also shown

that more research is needed to supply nurses with the knowledge and tools they need to

accomplish this task.

Conclusion

ESRD patients live in a limbo, suspended between life and death. Without the
life-sustaining technology of dialysis, they face certain death. Even with dialysis, their

continued survival is jeopardized by potential complications such as loss of dialysis

116

 

access and the worsening of comorbid conditions. ESRD patients’ daily lives are stressed
by the expectation of a significantly shortened life span and the constant threat of death.

Dialysis patients’ lives are also overshadowed by the stress of living each day
with a grave, incurable illness. End-stage renal disease places significant adaptive
demands on both patients and families. The treatment regimen, including dialysis, dietary
changes, and fluid restriction, requires permanent lifestyle changes that affect all aspects
of life. Daily life is difficult for both patients and families as a result.

Hemodialysis patients are especially appropriate recipients of palliative care
because of these challenges. Advanced care planning as well as other ethical,
psychosocial, and spiritual issues that arise due to limited life span have to be addressed.
Patients must also cope with threats to the quality of their daily living such as symptoms
and treatment side effects. Palliative care specializes in helping patients and families
manage these matters.

However, ESRD patients are not typically beneficiaries of these specialized
services. The need. for palliative care in, this population of patients has not been well
documented. A primary objective of this study was to provide the empirical evidence
necessary to justify access to palliative care, especially with respect to symptom
management. Several pertinent findings help to accomplish this goal, the most significant
being reports of unrelieved symptoms that negatively influence QOL. At a minimum,
these results focus attention on ESRD patients who are unable to access specialized
services that have helped to improve the lives of patients with other chronic illness such
as cancer, congestive heart failure, and chronic lung disease. This study is offered as a

Plea to healthcare providers, government officials, and private insurers, with the power to

117

 

 

remedy this injustice, to support efforts to improve the lives of this vulnerable and

forgotten group of patients through palliative care.

118

 

APPENDICES

119

Demographic Information

Subject Identification # Date

 

 

Day of week (circle one): M T W Th F S
1 2 3 4 5 6

Shift (check one):
7:00 AM (1) 12:00 PM (2) 5:00 PM (3)

1. Age years

 

2. What is the highest level of school you have completed?
(circle one number)

8th grade or less ................................... 1
Some high school ................................. 2
High school diploma or GED ................... 3
Vocational school or some college ............. 4
College degree .................................... 5
Professional or graduate degree ................ 6

3. What is your gender? (circle one number)
Male .......... 1
Female ....... 2

4. How do you describe yourself (circle one number)
Afiican American or Black ...................... 1
Hispanic or Latino ................................. 2
Native American or American Indian ........... 3
Asian or Pacific Islander .......................... 4
White, non Hispanic .............................. 5
Other (please specify) 6

 

5. What is your marital status?
(circle one number)

Married, living with husband/wife ................ 1
Single, never married ............................... 2
Divorced/separated .................................. 3
Widowed .............................................. 4

120

6. Do you have children?

(circle one number)
No ................. 0

Yes ................ l —’ If yes, how many children do you
have?

 

How many children live with you?

7. What is your current employment status?
(circle one number)

Employed full time .................................... 1
Employed part time ................................... 2
Retired .................................................. 3
Disabled ................................................ 4
Not employed for reasons unrelated to health. . 5
(circle one)
8. How long have you been on hemodialysis? months/years

 

9. Do you need help with personal care?

(circle one number)

 

No ............. 0
Yes ............ l _, If yes, who helps? spouse. 1
child .............. 2
fiiend ............. 3
paid assistant.... 4
other .............. 5
How many days/week does he/she help?
How many hours/day does he/she help?
10. Do you need help with household chores?
(circle one number)
No ............. 0
Yes ............ 1 _, If yes, who helps? spouse .......... 1
child ............ 2
friend............ 3
paid assistant... 4
other ............ 5

121

 

How many days/week does he/she help?

 

How many hours/day does he/she help?

 

11. Do you have any of the following health problems?

(circle all that apply)
No Yes

heart attack ............................. O 1
heart failure (fluid in lungs/heart

doesn’t pump well) ............ 0 1
clogged arteries in your legs ......... 0 1
numbness in feet/legs ................. 0 1
stroke .................................... O 1
chronic lung disease (emphysema,

bronchitis) ....................... 0 l
stomach ulcer/reflux disease ......... 0 1
diabetes ................................. 0 1
arthritis .................................. O 1
lupus ..................................... 0 1
fibromyalgia ............................ 0 l
cirrhosis of the liver ................... 0 1
cancer .................................... O l

1 2. Please list any other health problems that you have.

 

 

 

 

 

122

Chemistry Abnormality Score

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8.0 or > or 2.49 or <

 

Study Identification #
Check one: Current month Previous month
Laboratory Value Actual Value Criterion Values Points

K (mEq/L) 5.9 or < 0
6.0 or > 1
BUN (mg/dL) 100 or < 0
101 or > 1
Creatinine-female 14.9 or < 0
(mg/dL) 15.0 or > 1
Creatinine-male 19.9 or < 0
(mg/dL) 20.0 or > 1
Hct (%) 26 or > 0
20-25 1
19 or < 2
SGPT (U/L) 49 or < 0
50-200 1
201 or > 2
SGOT (U/L) 49 or < 0
50-200 1
201 or > 2
Alk Phos (U/L) 99 or < 0
100-200 1
201 or > 2
(Total Protein (G/dL) 6.0 or > 0
4.0-5.9 l
3.9 or < 2
Albumin (G/dL) 3.0 or > 0
2.0-2.9 l
1.9 or < 2
Triglycerides (mg/dL) < 2x normal limits 0
2x-3x normal limits 1
3x or more > normal limits 2
C02 (mEq/L) 20.1 or > 0
15-20 1
14.9 or < 2
Ca++ (mg/dL) 9.0-11.0 0
7.0-8.99 or 1101-130 1
m 6.99 or < or 13.01 or > 2
P04 = (mg/dL) 2.5-6 0
6.1-8.0 1
2

 

123

 

 

Symptoms and Level of Symptom Relief

Study Identification # Date:

 

 

Day of week completed: M T W Th F S

I am interested in learning about the symptoms you have. I have a list of 11 symptoms
that are commonly experienced by kidney patients. I will ask you if you have experienced
each of the symptoms in the past month. If so, I will ask you some further questions
about the symptom. I will also ask some questions about how you relieve the symptom
and the amount of relief you get.

124

 

Symptom #1 : Tiredness

1. Have you experienced tiredness in the past month?

El Yes If yes, answer questions below and on the next page.
E] No If no, go to symptom #2

A. How severe is the tiredness? (circle one)
1 2 3 4 5

not at all mild moderate severe very severe

B. On average, how often do you feel tired? (circle one)

1 2 3 4 5
not often 1-2 times every other most days every day
a week day

C. On average, how long does the tiredness last? (circle one)

1 2 3 4 5
not long quarter half of most of all day
of the day the day the day

D. How bothered are you by the tiredness? (circle one)

1 2 3 4 5
a little mildly moderately severely very severely
bothered bothered bothered bothered bothered

125

 

1. Ifyou experience tiredness, please place an X on the line next to ALL of the
methods you use to try to relieve it and answer questions A — D

Nothing

Take medicine. Name of medicine(s)
Rest/sleep/relaxation

Exercise

Distract yourself (for example, read, watch TV, listen to music)
Other

 

_—
_—
_—
_—
_—
_—

 

 

A. Of the methods checked above, please circle the one that you think is the
most effective in relieving the tiredness.

B. When you use the methods checked above to relieve the tiredness, how much
relief do you get? (circle one)

No relief 0 1 2 3 4 5 Total relief

C. How satisfied are you with the amount of relief you get from the tiredness?
(circle one)

Not at all 0 1 2 3 4 5 Completely
satisfied satisfied

D. How much do the methods checked above:
1) decrease the severity of the tiredness? (circle one)

Not at all 0 1 2 3 4 5 Very much

2) decrease how often you feel tired? (circle one)
Not at all 0 l 2 3 4 5 Very much

3) decrease how long the tiredness lasts? (circle one)
Not at all 0 l 2 3 4 5 Very much

4) make you less upset by the tiredness? (circle one)
Not at all 0 1 2 3 4 5 Very much

126

 

Symptom #2: Itching

2. Have you experienced itching in the past month?

[:1 Yes If yes, answer questions below and on the next page.
[:1 No Ifno, go to symptom #3

A. How severe is the itching? (circle one)
1 2 3 4 5

not at all mild moderate severe very severe

B. On average, how often do you itch? (circle one)

1 2 3 4 5
not often 1-2 times every other most days every day
a week day

C. On average, how long does the itching last? (circle one)

1 2 3 4 5
not long quarter half of most of all day
of the day the day the day

D. How bothered are you by the itching? (circle one)

1 2 3 4 5
a little mildly moderately severely very severely
bothered bothered bothered bothered bothered

127

2. If you experience itching, please place an X on the line next to ALL of the
methods you use to try to relieve it and answer questions A — D

Nothing

Take medicine. Name of medicine(s)
Rest/sleep/relaxation

Do skin care (for example, put on lotion, take a bath)

Distract yourself (for example, read, watch TV, listen to music)

Other

 

 

 

A. Of the methods checked above, please circle the one that you think is the
most effective in relieving the itching.

B. When you use the methods checked above to relieve the itching, how much
relief do you get? (circle one)

No relief 0 1 2 3 4 5 Total relief

C. How satisfied are you with the amount of relief you get from the itching?
(circle one)

Not at all 0 1 2 3 4 5 Completely
satisfied satisfied

D - How much do the methods checked above:
1) decrease the severity of the itching? (circle one)

Not at all 0 1 2 3 4 5 Very much

2) decrease how often you itch? (circle one)
Not at all 0 1 2 3 4 5 Very much

3) decrease how long the itching lasts? (circle one)
Not at all 0 1 2 3 4 5 Very much

4) make you less upset by the itching? (circle one)
Not at all 0 l 2 3 4 5 Very much

128

 

 

Symptom #3: Headaches

3. Have you experienced headaches in the past month?

 

El Yes If yes, answer questions below and on the next page.
El No If no, go to symptom #4

A. How severe are the headaches? (circle one)
1 2 3 4 5

not at all mild moderate severe very severe

B. On average, how often do you have a headache? (circle one)

1 2 3 4 5
not often 1-2 times every other most days every day
a week day

C. On average, how long does the headache last? (circle one)

1 2 3 4 5
not long quarter half of most of all day
of the day the day the day

D. How bothered are you by the headaches? (circle one)

1 2 3 4 5
a little mildly moderately severely very severely
bothered bothered bothered bothered bothered

129

 

3. If you experience headaches, please place an X on the line next to ALL of the
methods you use to try to relieve it and answer questions A — D

Nothing

Take medicine. Name of medicine(s)
Rest/sleep/relaxation

Massage

Distract yourself (for example, read, watch TV, listen to music)
Other

 

 

 

A. Of the methods checked above, please circle the one that you think is the
most effective in relieving the headaches.

B. When you use the methods checked above to relieve the headaches, how much
relief do you get? (circle one)

No relief 0 1 2 3 4 5 Total relief

C. How satisfied are you with the amount of relief you get from the headaches?
(circle one)

Not at all 0 1 2 3 4 5 Completely
satisfied satisfied

D . How much do the methods checked above:
1) decrease the severity of the headaches? (circle one)

Not at all 0 1 2 3 4 5 Very much

2) decrease how often you have headaches? (circle one)

Not at all 0 1 2 3 4 5 Very much

3) decrease how long the headaches lasts? (circle one)

Not at all 0 1 2 3 4 5 Very much

4) make you less upset by the headaches? (circle one)
Not at all 0 l 2 3 4 5 Very much

130

Symptom #4: Problems Sleeping

4. Have you experienced problems sleeping in the past month?

El Yes If yes, answer questions below and on the next page.
El No If no, go to symptom #5

A. How severe are the problems sleeping? (circle one)
1 2 3 4 5

not at all mild moderate severe very severe

B. On average, how often do you. have problems sleeping? (circle one)

1 2 3 4 5
not often 1-2 times every other most nights every night
a week night

C. On average, how long does it take you to fall asleep?

1 2 3 4 5
few minutes an hour awake most awake

not long
of the night all night

D . How bothered are you by the problems sleeping? (circle one)

1 2 3 4 5
a little mildly moderately severely very severely
bothered bothered bothered bothered bothered
13 1

 

 

4. If you experience problems sleeping, please place an X on the line next to
ALL of the methods you use to try to relieve it and answer questions A — D

Nothing
Take medicine. Name of medicine(s)
Rest/relaxation
. Cut out daytime naps
Distract yourself (for example, read, watch TV, listen to music)
Other

 

 

 

A. Of the methods checked above, please circle the one that you think is the
most effective in relieving your problems sleeping.

B. When you use the methods checked above to relieve your problems sleeping,
how much relief do you get? (circle one)

No relief 0 l 2 3 4 5 Total relief

C. How satisfied are you with the amount of relief you get from your problems
sleeping? (circle one)

Not at all 0 1 2 3 4 5 Completely
satisfied satisfied

D . How much do the methods checked above:
1) decrease the severity of your problems sleeping? (circle one)

Not at all 0 1 2 3 4 5 Very much

2) decrease how often you have problems sleeping? (circle one)

Not at all 0 1 2 3 4 5 Very much

3) decrease how long it takes you to get to sleep? (circle one)

Not at all 0 1 2 3 4 5 Very much

4) make you less upset by your sleeping problems? (circle one)

Not at all 0 1 2 3 4 5 Very much

132

 

Problem #5: Joint Pain

5. Have you experienced joint pain in the past month?

El Yes If yes, answer questions below and on the next page.
E] No If no, go to symptom #6

A. How severe is the joint pain? (circle one)
1 2 3 4 5

not at all mild moderate severe very severe

B. On average, how often do you have joint pain? (circle one)

1 2 3 4 5
not often 1-2 times every other most days every day
a week day

C. On average, how long does the joint pain last? (circle one)

1 2 3 4 5
not long quarter half of most of all day
of the day the day the day

D. How bothered are you by the joint pain? (circle one)

1 2 3 4 5
a little mildly moderately severely very severely
bothered bothered bothered bothered bothered

133

5. If you experience joint pain, please place an X on the line next to ALL of the
methods you use to try to relieve it and answer questions A — D

Nothing

Take medicine. Name of medicine(s)

Rest/limit activity/relaxation

Exercise

Distract yourself (for example, read, watch TV, listen to music)
Other

 

 

 

A. Of the methods checked above, please circle the one that you think is the
most effective in relieving the joint pain.

B. When you use the methods checked above to relieve the joint pain, how much
relief do you get? (circle one)

No relief 0 l 2 3 4 5 Total relief

C. How satisfied are you with the amount of relief you get from the joint pain?
(circle one)

Not at all 0 1 2 3 4 5 Completely
satisfied satisfied

D. How much do the methods checked above:
1) decrease the severity of the joint pain? (circle one)

Not at all 0 1 2 3 4 5 Very much

2) decrease how often you have joint pain? (circle one)

Not at all 0 1 2 3 4 5 Very much

3) decrease how long the joint pain lasts? (circle one)

Not at all 0 l 2 3 4 5 Very much

4) make you less upset by the joint pain? (circle one)

Not at all 0 1 2 3 4 5 Very much

134

Symptom #6: Cramps
6. Have you experienced cramps in the past month?

1:] Yes If yes, answer questions below and on the next page.
El No If no, go to symptom #7

A. How severe are the cramps? (circle one)
1 2 3 4 5

not at all mild moderate severe very severe

B. On average, how often do you have cramps? (circle one)

1 2 3 4 5
not often 1-2 times every other most days every day
a week day

C. On average, how long do the cramps last? (circle one)

1 2 3 4 5
not long quarter half of most of all day
of the day the day the day

D. How bothered are you by the cramps? (circle one)

1 2 3 4 5
a little mildly moderately severely very severely
bothered bothered bothered bothered bothered

135

6. If you experience cramps, please place an X on the line next to ALL of the
methods you use to try to relieve it and answer questions A — D

Nothing

Take medicine. Name of medicine(s)
Rest/limit activity/relaxation
Exercise

Distract yourself (for example, read, watch TV, listen to music)
Other

 

 

 

A. Of the methods checked above, please circle the one that you think is the
most effective in relieving the cramps.

B. When you use the methods checked above to relieve the cramps, how much
relief do you get? (circle one)

No relief 0 1 2 3 4 5 Total relief

C. How satisfied are you with the amount of relief you get from the cramps?
(circle one)

Not at all 0 1 2 3 4 5 Completely
satisfied satisfied

D. How much do the methods checked above:
1) decrease the severity of the cramps? (circle one)

Not at all 0 l 2 3 4 5 Very much

2) decrease how often you have cramps? (circle one)

Not at all 0 1 2 3 4 5 Very much

3) decrease how long the cramps lasts? (circle one)

Not at all 0 1 2 3 4 5 Very much

4) make you less upset by the cramps? (circle one)
Not at all 0 l 2 3 4 5 Very much

136

 

Symptom #7: Shortness of Breath
7. Have you experienced shortness of breath in the past month?

El Yes If yes, answer questions below and on the next page.
E] No If no, go to symptom 8

A. How severe is the shortness of breath? (circle one)
1 2 3 4 5

not at all mild moderate severe very severe

B. On average, how often do you feel short of breath? (circle one)

1 2 3 4 5
not often 1-2 times every other most days every day
a week day

C. On average, how long does the shortness of breath last? (circle one)

1 2 3 4 5
not long quarter half of most of all day
of the day the day the day

D. How bothered are you by the shortness of breath? (circle one)

1 2 3 4 5
a little mildly moderately severely very severely
bothered bothered bothered bothered bothered

137

7. If you experience shortness of breath, please place an X on the line next to
ALL of the methods you use to try to relieve it and answer questions A -- D

Nothing

Take medicine. Name of medicine(s)
Rest/sleep/relaxation

Change the way you are breathing (for example; purse your lips, deep
breathe)

Distract yourself (for example; read, watch TV, listen to music)
Other (for example; use fan, open window)

 

 

 

A. Of the methods checked above, please circle the one that you think is the
most effective in relieving the shortness of breath.

B. When you use the methods checked above to relieve the shortness of breath,
how much relief do you get? (circle one)

No relief 0 1 2 3 4 5 Total relief

C. How satisfied are you with the amount of relief you get from the shortness of
breath? (circle one)

Not at all 0 1 2 3 4 5 Completely
satisfied satisfied

D. How much do the methods checked above:
1) decrease the severity of the shortness of breath? (circle one)

Not at all 0 1 2 3 4 5 Very much

2) decrease how often you have shortness of breath? (circle one)

Not at all 0 1 2 3 4 5 Very much

3) decrease how long the shortness of breath lasts? (circle one)

Not at all 0 1 2 3 4 5 Very much

4) make you less upset by the shortness of breath? (circle one)

Not at all 0 1 2 3 4 5 Very much

138

Symptom #8: Chest Pain
8. Have you experienced chest pain in the past month?

El Yes If yes, answer questions below and on the next page.
C] No If no, go to symptom #9

A. How severe is the chest pain? (circle one)
1 2 3 4 5

not at all mild moderate severe very severe

B. On average, how often do you have chest pain? (circle one)

1 2 3 4 5
not often 1-2 times every other most days every day
a week day

C. On average, how long does the chest pain last? (circle one)

1 2 3 4 5
not long quarter half of most of all day
of the day the day the day

D. How bothered are you by the chest pain? (circle one)

1 2 3 4 5
a little mildly moderately severely very severely
bothered bothered bothered bothered bothered

139

. Ifyou experience chest pain, please place an X on the line next to ALL of the
methods you use to try to relieve it and answer questions A -— D

Nothing

Take medicine. Name of medicine(s)
Rest/sleep/relaxation

Change position

Distract yourself (for example, read, watch TV, listen to music)
Other

 

Illlll

 

 

. Of the methods checked above, please circle the one that you think is the
most effective in relieving the chest pain.

. When you use the methods checked above to relieve the chest pain, how much
relief do you get? (circle one)

No relief 0 1 2 3 4 5 Total relief

. How satisfied are you with the amount of relief you get from the chest pain?
(circle one)

Not at all 0 1 2 3 4 5 Completely
satisfied satisfied

. How much do the methods checked above:
1) decrease the severity of the chest pain? (circle one)

Not at all 0 l 2 3 4 5 Very much

2) decrease how often you have chest pain? (circle one)

Not at all 0 1 2 3 4 5 Very much

3) decrease how long the chest pain lasts? (circle one)

Not at all 0 l 2 3 4 5 Very much

4) make you less upset by the chest pain? (circle one)
Not at all 0 1 2 3 4 5 Very much

140

Symptom #9: Nausea and/or Vomiting

9. Have you experienced nausea/vomiting in the past month?

El Yes If yes, answer questions below and on the next page.
El No If no, go to symptom #10

A. How severe is the nausea/vomiting? (circle one)
1 2 3 4 5

not at all mild moderate severe very severe

B. On average, how often do you have nausea/vomiting? (circle one)

1 2 3 4 5
not often 1-2 times every other most days every day
a week day

C. On average, how long does the nausea/vomiting last? (circle one)

1 2 3 4 5
not long quarter half of most of all day
of the day the day the day

D. How bothered are you by the nausea/vomiting? (circle one)

1 2 3 4 5
a little mildly moderately severely very severely
bothered bothered bothered bothered bothered

141

9. If you experience nausea/vomiting, please place an X on the line next to ALL
of the methods you use to try to relieve it and answer questions A — D

Nothing

Take medicine. Name of medicine(s)

Rest/sleep/relaxation

Change eating pattern (for example; eat small meals often)

Distract yourself (for example, read, watch TV, listen to music)
Other

 

 

 

A. Of the methods checked above, please circle the one that you think is the
most effective in relieving the nausea/vomiting.

B. When you use the methods checked above to relieve the nausea/vomiting,
how much relief do you get? (circle one)

No relief 0 l 2 3 4 5 Total relief

C. How satisfied are you with the amount of relief you get from the
nausea/vomiting? (circle one)

Not at all 0 1 2 3 4 5 Completely
satisfied - satisfied.

D. How much do the methods checked above:
1) decrease the severity of the nausea/vomiting? (circle one)

Not at all 0 1 2 3 4 5 Very much

2) decrease how often you have nausea/vomiting? (circle one)

Not at all 0 1 2 3 4 5 Very much

3) decrease how long the nausea/vomiting lasts? (circle one)

Not at all 0 1 2 3 4 5 Very much

4) make you less upset by the nausea/vomiting? (circle one)
Not at all 0 1 2 3 4 5 Very much

142

Symptom #10: Abdominal Pain
10. Have you experienced abdominal pain in the past month?

Cl Yes If yes, answer questions below and on the next page.
El No If no, go to symptom 11

A. How severe is the abdominal pain? (circle one)
1 2 3 4 5

not at all mild moderate severe very severe

B. On average, how often do you have abdominal pain? (circle one)

1 2 3 4 5
not often 1-2 times every other most days every day
a week day

C. On average, how long does the abdominal pain last? (circle one)

1 2 3 4 5
not long quarter half of most of all day
of the day the day the day

D. How bothered are you by the abdominal pain? (circle one)

1 2 3 4 5
a little mildly moderately severely very severely
bothered bothered bothered bothered bothered

143

 

10. If you experience abdominal pain, please place an X on the line next to ALL
of the methods you use to try to relieve it and answer questions A — D

Nothing
Take medicine. Name of medicine(s)
Rest/sleep/relaxation
Change position
Distract yourself (for example, read, watch TV, listen to music)
Other

 

 

 

A. Of the methods checked above, please circle the one that you think is the
most effective in relieving the abdominal pain.

B. When you use the methods checked above to relieve the abdominal pain, how
much relief do you get? (circle one)

No relief 0 l 2 3 4 5 Total relief

C. How satisfied are you with the amount of relief you get from the abdominal
pain? (circle one)

Not at all 0 1 2 3 4 5 Completely
satisfied satisfied

D. How much do the methods checked above:
1) decrease the severity of the abdominal pain? (circle one)

Not at all 0 1 2 3 4 5 Very much

2) decrease how often you have abdominal pain? (circle one)

Not at all 0 1 2 3 4 5 Very much

3) decrease how long the abdominal pain lasts? (circle one)
Not at all 0 1 2 3 4 5 Very much

4) make you less upset by the abdominal pain? (circle one)
Not at all 0 1 2 3 4 5 Very much

144

Symptom #11: Muscle Weakness
11. Have you experienced muscle weakness in the past month?

El Yes If yes, answer questions below and on the next page.
El No If no, go to the last page.

A. How severe is the muscle weakness? (circle one)
1 2 3 4 5

not at all mild moderate severe very severe

B. On average, how often do you have muscle weakness? (circle one)

1 2 3 4 5
not often 1-2 times every other most days every day
a week day

C. On average, how long does the muscle weakness last? (circle one)

1 2 3 4 5
not long quarter half of most of all day
of the day the day the day

D. How bothered are you by the muscle weakness? (circle one)

1 2 3 4 5
a little mildly moderately severely very severely
bothered bothered bothered bothered bothered

145

 

11. If you experience muscle weakness, please place an X on the line next to ALL
of the methods you use to try to relieve it and answer questions A — D

Nothing
Take medicine. Name of medicine(s)
Rest/sleep/relaxation
Exercise
Distract yourself (for example, read, watch TV, listen to music)
Other

 

 

 

A. Of the methods checked above, please circle the one that you think is the
most effective in relieving the muscle weakness.

B. When you use the methods checked above to relieve the muscle weakness,
how much relief do you get? (circle one)

No relief 0 1 2 3 4 5 Total relief

C. How satisfied are you with the amount of relief you get from the muscle
weakness? (circle one)

Not at all 0 l 2 3 4 5 Completely
satisfied satisfied

D. How much do the methods checked above:
1) decrease the severity of the muscle weakness? (circle one)

Not at all 0 1 2 3 4 5 Very much

2) decrease how often you have muscle weakness? (circle one)

Not at all 0 l 2 3 4 5 Very much

3) decrease how long the muscle weakness lasts? (circle one)

Not at all 0 1 2 3 4 5 Very much

4) make you less upset by the muscle weakness? (circle one)

Not at all 0 1 2 3 4 5 Very much

146

 

12. Are there any other symptoms that you experienced in the past month? If so, list
them below along with what you did to get relief and how effective it was.

 

Symptom

Method of Relief

Effectiveness
0 (not at all effective)
5 (completely effective)

 

 

 

 

 

 

 

 

147

 

 

Study Identification # Date

 

 

Day of week (circle one): M T W TH F S

Besides learning about your physical symptoms, I am also interested in learning how you
feel emotionally. I will read you a question and four responses. Tell me which response
best fits how you have been feeling in the past week.

1. I feel tense or ‘wound up’:
A. most of the time
B. a lot of the time
C. from time to time, occasionally
D. not at all

2. I still enjoy the things I used to enjoy:
A. definitely as much
B. not quite so much
C. only a little
D. hardly at all

3. I get a sort of frightened feeling as if something awful is about to happen:
very definitely and quite badly

yes, but not too badly

a little, but it doesn’t worry me

not at all

.5095?

4. I can laugh and see the funny side of things:
A. as much as I always could
B. not quite so much now
C. definitely not so much now
D. not at all

5. Worrying thoughts go through my mind:
A. a great deal of the time
B. a lot of the time
C. from time to time but not too often
D. only occasionally

148

6. I feel cheerful:

not at all

not often
sometimes

most of the time

new»

7. I can sit at ease and feel relaxed:
A. definitely
B. usually
C. not often
D. not at all

8. I feel as if I am slowed down:
A. nearly all of the time
B. very often
C. sometimes
D. not at all

9. I get assort of frightened feeling like ‘butterflies’ in the stomach:
not at all
occasionally
quite often
very often

POP”?

10. I have lost interest in my appearance:
A. definitely
B. I don’t take so much care as I should
C. I may not take quite as much care
D. I take just as much care as ever

1 1. I feel restless as if I have to be on the move:
A. very much indeed
B. quite a lot
C. not very much
D. not at all

149

 

12. I look forward with enjoyment to things:
A. as much as I ever did
B. rather less than I used to
C. definitely less than I used to
D. hardly at all

13. I get sudden feelings of panic:
very often indeed

quite often

not very often

not at all

POP”?

14. I can enjoy a good book or radio or TV program:
A. often
B. sometimes
C. not often
D. very seldom

150

 

Ferrans and Powers
QUALITY OF LIFE INDEX©
DIALYSIS VERSION - III

PART 1. For each of the following, please choose the answer that best describes how
satisfied you are with that area of your life. Please mark your answer by circling the
number. There are no right or wrong answers.

1 Very dissatisfied

2 Moderately dissatisfied
3 Slightly dissatisfied

4 Slightly satisfied

5 Moderately satisfied

6 Very satisfied

HOW SATISFIED ARE YOU WITH:

 

1. Your health? 1 2 3 4 5

 

2. Your health care? 1 2 3 4 5

 

3. The amount of energy you have for everyday
activities? 1 2 3 4 5

 

4. Your ability to take care of yourself without

 

 

 

 

 

 

 

 

help? 1 2 3 4 5
5. The likelihood you will get a kidney transplant? 1 2 3 4 5
6. The changes you have had to make in your life

because of kidney failure (such as diet and the

need for dialysis)? 1 2 3 4 5
7. The amount of control you have over your life? 1 2 3 4 5
8. Your chances of living as long as you would

like? 1 2 3 4 5
9. Your family’s health? 1 2 3 4 5
10. Your children? 1 2 3 4 5
11. Your family’s happiness 1 2 3 4 5
12. Your sex life? 1 2 3 4 5

 

151

 

13.

Your spouse, lover, or partner?

 

14.

The emotional support you get from your
family?

 

15.

Your friends?

 

16.

The emotional support you get from people
other than your family?

 

17.

Your ability to take care of family
responsibilities?

 

18.

How useful you are to other people?

 

19.

The amount of worries in your life?

 

20.

Your neighborhood?

 

21.

Your home, apartment, or place where you
live?

 

22.

Your job (if employed)?

 

23.

Not having a job (if unemployed, retire, or
disables)?

 

24.

Your education?

 

25.

How well you can take care of your financial
needs?

 

26.

The things you do for fun?

 

27.

Your chances for a happy future?

 

28.

Your peace of mind?

 

29.

Your faith in God?

 

30.

Your achievement of personal goals?

 

31.

Your happiness in general?

 

152

 

32. Your life in general? 1 2 3 4 5 6

 

33. Your personal appearance? 1 2 3 4 5 6

 

34. Yourself in general? 1 2 3 4 5 6

 

PART 2. For each of the following, please choose the answer that best describes how
important that area of your life is to you. Please mark your answer by circling the number.
There are no right or wrong answers.

 

 

HOW IMPORTANT TO YOU 18:
1. Your health? 1 2 3 4 5 6
2. Your health care? 1 2 3 4 5 6

 

3. The amount of energy you have for everyday
activities? 1 2 3 4 5 6

 

4. Your ability to take care of yourself without

 

 

 

 

 

 

 

 

 

help? 1 2 3 4 5 6
5. The likelihood you will get a kidney transplant? 1 2 3 4 5 6
6. The changes you have had to make in your life

because of kidney failure (such as diet and the

need for dialysis)? 1 2 3 4 5 6
7. The amount of control you have over your life? 1 2 3 4 5 6
8. Your chances of living as long as you would

like? 1 2 3 4 5 6
9. Your family’s health? 1 2 3 4 5 6
10. Your children? 1 2 3 4 5 6
11. Your family’s happiness 1 2 3 4 5 6
12. Your sex life? 1 2 3 4 5 6
13. Your spouse, lover, or partner? 1 2 3 4 5 6

 

153

 

 

14. The emotional support you get from your

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

family? 1 2 3 4 5 6
15. Your friends? 1 2 3 4 5 6
16. The emotional support you get from people

other than your family? 1 2 3 4 5 6
17. Your ability to take care of family

responsibilities? 1 2 3 4 5 6
18. How useful you are to other people? 1 2 3 4 5 6
19. The amount of worries in your life? 1 2 3 4 5 6
20. Your neighborhood? 1 2 3 4 5 6
21. Your home, apartment, or place where you

live? 1 2 3 4 5 6
22. Your job (if employed)? 1 2 3 4 5 6
23. Not having a job (if unemployed, retire, or

disables)? l 2 3 4 5 6
24. Your education? 1 2 3 4 5 6
25. How well you can take care of your financial

needs? 1 2 3 4 5 6
26. The things you do for fun? 1 2 3 4 5 6
27. Your chances for a happy fiiture? 1 2 3 4 5 6
28. Your peace of mind? 1 2 3 4 5 6
29. Your faith in God? 1 2 3 4 5 6
30. Your achievement of personal goals? 1 2 3 4 5 6
31. Your happiness in general? 1 2 3 4 5 6
32. Your life in general? 1 2 3 4 5 6

 

154

 

33. Your personal appearance? 1 2 3 4 5

 

34. Yourself in general? 1 2 3 4 5

 

© Copyright 1984 & 1998 Carol Estwing Ferrans and Marjorie J. Powers (Do not use
without permission).

155

REFERENCES

Aiken, L. S., & West, 8. G. (1991). Multiple regression: Testing and interpreting
interactions. Newbury Park: Sage Publications.

Anderson, K. N., Anderson, L. E., & Glanze, W. D. (Eds). (1998). Mosby ’s medical,
nursing, and allied health dictionary. New York: Mosby—Year Books, Inc.

Aylard, P. R., Gooding, J. H., McKenna, P. J ., & Snaith, R. P. (1987). A validation study
of three anxiety and depression self-assessment scales. Journal of Psychosomatic
Research, 31, 261-268.

Barrett, B., Vavasour, H., Major, A., & Parfrey, P. (1990). Clinical and psychological
correlates of somatic symptoms in patients on dialysis. Nephron, 55, 10-15.

Barriers to hospice care: are we shortchanging dying patients: hearing before the Special
Committee on Aging, United States Senate, 106d Cong., 1, 2000.

Brown, J. W. (2001). Techniques of muiltivariate analysis. Factor analysis. Retrieved on
March 15, 2004, from http://ocean.otr.usm.edu/~jwbrown/chapter_five.htm.

Bruni, A., Dos-Ramos-Farias, E., Heguilen, R., Galimberti, R., Laghens, M., Petrolito, J .,
et al. (1989). Pruritis in chronic kidney insufficiency. Abstract obtained from
SilverPlatter Medline 1989-1992.

Brunier, G., & Graydon, J. (1992). The relationship of anemia, nonspecific uremic
symptoms, and physical activity to fatigue in patients with end stage renal disease
on hemodialysis. Abstract obtained from American Nephrology Nurses
Association Journal, 19, 157.

Cardenas, D., & Kutner, N. (1982). The problem of fatigue in dialysis patients. Nephron,
30, 336-340.

Carpenito, L. J. (1997). Nursing diagnosis. Application to clinical practice. (7th ed.). New
York: Lippincott.

' Charlson, M. E., Pompei, P., Alex, K. L., & MacKenzie, C. R. (1987). A new method of
classifying prognostic comorbidity in longitudinal studies: Development and
validation. Journal of Chronic Diseases, 40, 373-383.

Christensen, A., & Ehlers, S. (2002). Psychological factors in end-stage renal disease: An
emerging context for behavioral medicine research. Journal of Consulting and

Clinical Psychology, 70, 712-724.

CLIA Program; Approval of the College of American Pathologists, 60 Fed. Reg. 7,774
(Feb. 9, 1995).

156

College of American Pathologists. (2004). About CAP laboratory accreditation
programs. Retrieved on February 11, 2004, from http://www.cap.org/apps/
cap.portal?_n1pb=true&_pageLabel=lab_accred_book.

Colorado HealthSite. (2004). A guide to blood chemistries for kidney dialysis patients.
Retrieved on January 13, 2004, from http:www.coloradohealthsite.org/
dialysis/bc/bc20.htm.

Corbin, J. M., & Strauss, A. (1992). A nursing model for chronic illness management
based upon the trajectory framework. In P. Woog (Ed.), The chronic illness
trajectory framework. The Corbin and Strauss nursing model. New York:
Springer Publishing Company.

Craven, J ., Littlefield, C., Rodin, G., & Murray, M. (1991). The Endstage Renal Disease
Severity Index (ESRD-SI). Psychological Medicine, 21, 237—243.

Curtin, R., Bultrnan, D., & Thomas-Hawkins, C. (2002). Hemodialysis patients’
symptom experiences: Effects on physical and mental functioning. Nephrology
Nursing Journal, 29, 562-574.

Curtis, J. (2004). Daily short and nightly nocturnal home hemodialysis: State of the art.
Dialysis and Transplantation, 23(3), 64-73.

Custer, R. L., Scarcella, J. A., & Stewart, B. R. (1999). The modified Delphi technique-A
rotational modification. Journal of Vocational and Technical Education, 15(2).
Retrieved December 22, 2003, from http://scholar.lib.vt.edu/ejoumals /JVTE/

v1 5n2/custer.html.

Daugirdas, J., & Ing, T. (1994). Handbook ofdialysis (2nd ed.). New York: Little, Brown
and Company.

Deshotels, A., Planchock, N., Dech, Z., & Prevost, S. (1995). Gender differences in
perceptions of quality of life in cardiac rehabilitation patients. Journal of
Cardiopulmonary Rehabilitation, 15(2), 143-148.

Dodd, M. J ., J anson, S., F acione, N., Faucett, J ., Froelicher, E. S., Humphreys, J ., et al.
(2001a). Advancing the science of symptom management. Journal of Advanced
Nursing, 33, 668-676.

Dodd, M. J ., Miaskowski, C., & Paul, S. (2001b). Symptom clusters and their effect on
the functional status of patients with cancer. Oncology Nursing Forum, 28,
465-470.

Factor analysis using SAS proc factor. (1995, June). Retrieved on November 19, 2003,
fi'om http://www.utexas.edu/cc/docs/stat53.htrnl.

157

Ferrans, C. E. (1997). Development of a conceptual model of quality of life. In A. Gift
(Ed.), Clarifying concepts in nursing research (pp. 110-121). New York: Springer
Publishing Company.

F errans, C. E. (1990). Development of the Quality of Life Index for patients with
cancer. Oncology Nursing Forum, 17(Suppl. 3), 15-19, 20-21.

F errans, C. E. (2004). Ferrans and Powers Quality of Life Index. Retrieved on February
12, 2004, from http://www.uic.edu/orgs/qli/.

Ferrans, C. E., & Powers, M. J. (1992). Psychometric assessment of the Quality of Life
Index. Research in Nursing and Health, 15, 29-38.

Ferrans, C. E., & Powers, M. J. (1985). Quality of Life Index: Development and
psychometric properties. Advances in Nursing Science, 8, 15-24.

F errans, C. E., & Powers, M. J. (1993). Quality of life of hemodialysis patients. American
Nephrology Nurses Association Journal, 20, 575-581.

Ferrell, B. R., & Coyle, N. (Eds). (2001). Textbook of palliative nursing. New York:
Oxford University Press.

Flint, A. J ., & Rifat, S. L. (2002). Factor structure of the Hospital Anxiety and
Depression Scale in older patients with major depression. International Journal of
Geriatric Psychiatry, 1 7(2), 117-123.

George, D., & Mallery, P. (2003). SPSS for windows step by step 11.0 update (4th ed.).
New York: Allyn and Bacon.

Gift, A., J ablonski, A., Stommel, M., & Given, C. W. (2004). Symptom clusters in
elderly patients with lung cancer. Oncology Nursing Forum, 31, 202-212.

Gift, A., Stommel, M., Jablonski, A., & Given, C. W. (2003). A cluster of symptoms over
time in patients with lung cancer. Nursing Research, 52, 393-400.

Good Samaritan Hospital Chronic Dialysis Center. (2004). Blood lab value comparison
normal (non-ESRD) to acceptable ESRD normal. Retrieved on February 13,
2004, from http://www.azstamet.com/~bsmith/tools/labvalus.htm.

Griffin, K. W., Friend, R., & Wadhwa, N. K. (1995). Measuring disease severity in
patients with end-stage renal disease: Validity of the Craven et al. ESRD Severity
Index. Psychological Medicine, 25, 189-193.

Hamilton, G., & Locking-Cusolito, H. (1998). The relationship between dialysis
adequacy and quality of life: A report of a pilot study. CANNT, 8(3), 25-29.

158

Hoffart, N. (1995). Characteristics of nephrology nurse researchers and their research.
American Nephrology Nurses Association Journal, 22, 333-341.

Jablonski, A. (2004). The illness trajectory of end stage renal disease dialysis patients.
Research and Theory in Nursing Practice: An International Journal, 18, 51-72.

J ablonski, A., Gift, A., Von Eye, A., Given, C. W., & Given, B. (2004). A test of the
Theory of Unpleasant Symptoms. Manuscript submitted for publication.

Jakic, M. (1999). Does uremic pruritus in hemodialyzed patients disappear only with
replacement therapy. Abstract obtained from SilverPlatter Medline1999-2000.

Katz, J. N., Chang, L. C., Sangha, O., Fossel, A. H., & Bates, D. W. (1996). Can
comorbidity be measured by questionnaire rather than medical record review?
Medical Care, 34(1), 73-84.

Khan, 1. H. (1998). Comorbidity: The major challenge for survival and quality of life in
end-stage renal disease. Nephrology Dialysis Transplantation, 13(Suppl 1), 76-79.

Killingworth, A., & Van Den Akker, O. (1996). The quality of life of renal dialysis
patients: Trying to find the missing measurement. International Journal of
Nursing Studies, 33, 107-120.

Kimmel, P. (2002). Depression in patients with chronic renal disease. Journal of
Psychosomatic Research, 53, 951-956.

Kimmel, P., Peterson, R., Weihs, K. Simmens, S., Boyle, D., Cruz, 1., et al. (1995).
Aspects of quality of life in hemodialysis patients. Journal of the American
Society of Nephrology, 6, 1418-1426.

Kristjanson, L. J. (2001). Establishing goals: Communication traps and treatment lane
changes. In B. R. Ferrell & N. Coyle (Eds), Textbook of palliative nursing
(pp. 331-338). New York: Oxford University Press.

Kurtz, M. E., Kurtz, J. C., Stommel, M., Given, C.W., & Given, B. A. (2002). Predictors
of depressive symptomatology of geriatric patients with lung cancer—~—a
longitudinal analysis. Psycho-Oncology, 11, 12-22.

Kutner, N., Brogan, D., Fielding, B., & Hall, W. (2000). Race and symptoms in older
hemodialysis patients. Ethnicity & Disease, 10, 328-333.

Lenz, E. R., Pugh, L. C., Milligan, R. A., Gift, A., & Suppe, F. (1997). The middle-range
Theory of Unpleasant Symptoms: An update. Advances in Nursing Science,
19(3), 14-27.

Lindsay, R., Kortas, C., & the Daily/Noctumal Dialysis Study Group. (2001). Hemeral

159

(daily) hemodialysis. Advances in Renal Replacement Therapy, 8, 236-249.

Lye, W. C., Chan, P., Leong, S. O., & van der Straaten, C. (1997). Psychosocial and
psychiatric morbidity in patients on CAPD. Advances in Peritoneal Dialysis, 13,
134-136.

Maduell, F ., Navarro, V., Torregrosa, E., Ruis, A., Dicenta, F., Cruz, M., et al. (2003).
Change from three times a week on-line hemodiafiltration to short daily on-line
hemodiafiltration. Kidney International, 64, 305-313.

McCann, K., & Boore, J. (2000). Fatigue in persons with renal failure who require
maintenance haemodialysis. Journal of Advanced Nursing, 32, 1132-1142.

McDaniel, R. W., & Rhodes, V. A. (1995). Symptom experience. Seminars in Oncology
Nursing, 11, 232-234.

Merkus, M., Jager, K., Dekker, F., de Haan, R., & Boeschoten, E. (1999). Physical
symptoms and quality of life in patients on chronic dialysis: Results of the
Netherlands Cooperative Study on Adequacy of Dialysis (NECOSAD).
Nephrology Dialysis Transplantation, I4, 1 163-1 170.

Meyer, K., Espindle, D., DeGiacomo, J ., J enuleson, C., Kurtin, P., & Davies, A. (1994).
Monitoring dialysis patients’ health status. American Journal of Kidney Diseases,
24, 267-279.

Miller, L., Nail, L., Rosenfeld, A., & Perrin, N. (2004, April). Symptom cluster research.
Symposium at the meeting of the Western Institute of Nursing 37th Annual
Communicating Nursing Research Conference, Portland, OR.

Moorey, S., Greer, S., Watson, M., Gonnan, C., Rowden, L., Tunmore, R., et al. (1991).
The factor structure and factor stability of the Hospital Anxiety and Depression
Scale in patients with cancer. British Journal of Psychiatry, 158, 255-259.

Morton, A. R., Meers, C., Singer, M., Toffelrnire, E., Hopman, W., Comb, J ., et al.
(1996). Quantity of dialysis: Quality of life—what is the relationship? ASAIO
Journal, 42, M7l3—M717.

Moss, A., & the ESRD Workgroup. (2001). End-stage renal disease workgroup full
report. Recommendations to the field. Missoula, MT: The Robert Wood Johnson

Foundation.

National Institute of Diabetes and Digestive and Kidney Disases. (2001). Hemodialysis
dose and adequacy. Retrieved on August 20, 2003, from
http://kidney.niddk.nih.gov/kidneydiseases/pubs/hemodialysisdose/index.htm.

National Kidney Foundation. (2001). Clinical practice guidelines for hemodialysis

160

 

 

adequacy. Retrieved on February 6, 2004, from http://www.kidney.org/
professionals/doqi/guidelines/doqiuphd_i.html.

National Kidney Foundation. (1997). National Kidney Foundation Dialysis Outcomes
Quality Initiative. Clinical practice guidelines for hemodialysis adequacy.
Retrieved on March 15, 2004, from http://www.kidney.org/professionals/
kdoqi/guidelines_updates/doqiuphd_intro.htrnl.

Nelson, A. J. (2002). Using a modified Delphi methodology to develop a competency
model for vet practitioners. Retrieved on December 22, 2003, from
http://home.earthlink.net/~amkefossen/AJN/publications/RMS02EPaper.pdf.

Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric theory (3rd ed.). New York:
McGraw-Hill, Inc.

Oleson, M. (1990). Content validity of the Quality of Life Index. Applied Nursing
Research, 3, 126-127.

Parfrey, P. S., Vavasour, H. M., Bullock, M., Henry, S., Hamett, J. D., & Gault, M. H.
(1989). Development of a health questionnaire specific for end-stage renal
disease. Nephron, 52, 20-28.

Parfrey, P. S., Vavasour, H. M., Bullock, S., Henry, S., Hamett, J. D., & Gault, M. H.

(1987). Symptoms in end-stage renal disease: Dialysis v transplantation.
Transplantation Proceedings, 19, 3407-3409.

Parfrey, P. S., Vavasour, H., Henry, S., Bullock, M., & Gault, M. H. (1988). Clinical
features and severity of nonspecific symptoms in dialysis patients. Nephron, 50,
121-129.

Poppel, D. M., Cohen, L. M., & Germain, M. J. (2001). The renal palliative care
initiative. Unpublished manuscript.

Reikes, S. T. (2000). Trends in end-stage renal disease: Epidemiology, morbidity, and
mortality. Postgraduate Medicine, 108(1), 124-142.

Renal Network of the Upper Midwest, Inc. (2001). End Stage Renal Disease Network 11
2000 annual report. St. Paul, MN: Author.

Rettig, R., & Levinsky, N. (1991). Kidney failure and the federal government.
Washington, DC: National Academy Press.

Rhodes, V. A., & Watson, P. M. (1987). Symptom distress—the concept: Past and
present. Seminars in Oncology Nursing, 3, 242-247.

Rolland, J. S. (1987). Chronic illness and the life cycle: A conceptual framework. Family

161

Process, 26, 203-221.

Simon, G. (2001). Treating depression in patients with chronic disease. Western Journal
of Medicine, 1 75, 292-293.

Sloan, R. S. (1999). Living a life-sustained-by—medical-technology: Dialysis is killing
me. Paper presented at the 23rd Midwest Nursing Research Society Conference,
Indianapolis, IN.

Snaith, R. P. (2003). The Hospital Anxiety and Depression Scale. Retrieved on March 15,
2004, from http://www.pubmedcentral.nih.gov/articlerender.fcgi?articl=183848.

Snaith, R. P., & Zigmond, A. S. (1994). Hospital Anxiety and Depression Scale.
Retrieved on March 15, 2004, from http://www.nfer-nelson.co.uk/
catalogue/catalogue_detail.asp?catid+98&id=l 125.

Srivastava, R. H. (1989). Fatigue in end-stage renal disease patients. In S. Funk, E.
Tomquist, M. Champagne, A. Copp, & R. Weise (Eds), The management of pain,
fatigue, and nausea (pp. 217-224). New York: Springer Publishing Company.

Stevens, J. S. (2002). Applied multivariate statistics for the social sciences (4th ed.).
Mahwah, NJ: Lawrence Erlbaum Associates.

Stone, A., Shiffrnan, S., Schwartz, J ., Broderick, J., & Hufford, M. (2002). Patient
non-compliance with paper diaries. British Medical Journal, 324, 1193-1194.

The University of California, San Francisco School of Nursing Symptom Management
Faculty Group. (1994). A model for symptom management. Image: Journal of
Nursing Scholarship, 26, 272-276.

U. S. Department of Health and Human Services. (2004). End stage renal disease
medical evidence report Medicare entitlement and/or patient registration.
Retrieved March 15, 2004, fi'om http://www.cms.hhs.gov/esrd/5a,pdf.

U. S. Department of Health and Human Services. (2003). Protecting personal health
information in research: Understanding the HIPPA Privacy Rule. Retrieved
March 15, 2004, from http://privacyruleandresearch.nih.gov/pr__02.asp.

U. S. Department of Health and Human Services Health Care Financing. (2003).
Medicare hospice benefits. Retrieved March 15, 2004, from
http://www.medicare.gov/Publications/Pubs/pdf/02154.pdf.

United States Renal Data System. (2002). 2002 Annual data report: Atlas of end-stage

renal disease in the United States. Bethesda, MD: National Institutes of Health,
National Institute of Diabetes and Digestive and Kidney Diseases.

162

Vachon, M. (2001). The nurse’s role: The world of palliative care nursing. In B. R.
Ferrell & N. Coyle (Eds), Textbook of palliative nursing (pp. 647-662). New
York: Oxford University Press.

Vanholder, R., Glorieux, G., De Smet, R., & De Deyn, P. (2004). Low water-soluble
uremic toxins. Advances in Renal Replacement Therapy, 10, 257-269.

Walker, 3., Fine, A., & Kryger, M. (1995). Sleep complaints are common in a dialysis
unit. American Journal of Kidney Diseases, 26, 751-756.

Watts, S. C., Bhutani, G. E., Stout, I. H., Ducker, C. M., Cleator, P. J., McGarry, J., et al.
(2002). Mental health in older adult recipients of primary care services: Is
depression the key issue? Identification, treatment, and the general practitioner.
International Journal of Geriatric Psychiatry, 1 7, 427-437.

Wilson, I. B., & Cleary, P. D. (1995). Linking clinical variables with health-related
quality of life. A conceptual model of patient outcomes. Journal of the American
Medical Association, 273, 59-65.

Wolcott, D. L., & Nissenson, A. R. (1988). Quality of life in chronic dialysis patients: A
critical comparison of continuous ambulatory peritoneal dialysis (CAPD) and
hemodialysis. American Journal of Kidney Diseases, 11, 402-412.

Wolcott, D. L., Nissenson, A. R., & Landsverk, J. (1988). Quality of life in chronic
dialysis patients factors unrelated to dialysis modality. General Hospital
Psychiatry, 10, 267-277.

World Health Organization. (2001). WHO definition of palliative care. Retrieved March
15, 2004, from http://www.who.int/cancer/palliative/definition/en/.

Zigmond, A. S., & Snaith, R. P. (1983). The Hospital Anxiety and Depression Scale.
Acta Psychiatrica Scandinavica, 67, 361-370.

163

 

 

 

 

gum:

"em

 

.. , 31..

 

Enr- a»:

m

warm»: