USF Libraries
USF Digital Collections

Relationship between nurses' management of pediatric oncology patients' symptoms and job satisfaction

MISSING IMAGE

Material Information

Title:
Relationship between nurses' management of pediatric oncology patients' symptoms and job satisfaction
Physical Description:
Book
Language:
English
Creator:
Rheingans, Jennifer I
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
Children
Cancer
Distress
Job satisfaction
Nursing interventions
Dissertations, Academic -- Nursing -- Doctoral -- USF   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: A primary function of the pediatric oncology nurse is to provide symptom management to children with cancer. Symptom management strategies have been published, but there is scarce literature examining neither the actual use of these nursing interventions, nor the effects of using these interventions on the nurses' perceived work environment. The purpose of this study was to examine the nursing interventions used in treating pediatric oncology patients' symptoms, as well as the emotional sequelae from providing this care. Phase One of this study examined the content validity of the newly developed Nurses Distress and Interventions for Symptoms Survey (NDISS) utilizing content experts. Phase Two of this study involved both the reliability testing of the NDISS by test-retest and served as a pilot for Phase Three.^ In Phase Three, a national sample of pediatric oncology nurses was surveyed about their patients' symptoms, the nurses' distress from the symptoms, the nursing interventions used to treat the symptoms, the perceived efficacy of the nursing interventions, and their job satisfaction. The response rate was 53%, and analysis of study hypotheses were evaluated using Pearson's correlation and multiple regression analyses. The main study variables were not related in the hypothesized direction; therefore four of the six hypotheses were not supported. However, quantity and perceived effectiveness of nursing interventions were both found to act as mediators in the study model, and as a result, these two hypotheses were retained.^ The results of the survey demonstrated a high frequency of distressing patient symptoms as perceived by nurses (mean 6, range 0-7); nurses rated their distress from these symptoms as moderate (mean 2.9, range 0-4); nurses used an average of 12.7 nursing interventions per symptom (range 0-38); nurses found the nursing interventions moderately effective (mean 2.5, range 0-4); and nurses had moderately high overall job satisfaction (mean 3.9, range 1-5). Although many of the hypotheses were not supported, interesting trends in the data were found. In addition, the findings provided elucidation of specific nursing interventions used by pediatric oncology nurses as well as a description of the effects of providing patients' symptom management, including nurses' distress, perceived effectiveness of nursing interventions, and job satisfaction.
Thesis:
Dissertation (Ph.D.)--University of South Florida, 2007.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Jennifer I. Rheingans.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 90 pages.
General Note:
Includes vita.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001968111
oclc - 268954259
usfldc doi - E14-SFE0002199
usfldc handle - e14.2199
System ID:
SFS0026517:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001968111
003 fts
005 20081105100429.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 081105s2007 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002199
035
(OCoLC)268954259
040
FHM
c FHM
049
FHMM
090
RT41 (ONLINE)
1 100
Rheingans, Jennifer I.
0 245
Relationship between nurses' management of pediatric oncology patients' symptoms and job satisfaction
h [electronic resource] /
by Jennifer I. Rheingans.
260
[Tampa, Fla.] :
b University of South Florida,
2007.
3 520
ABSTRACT: A primary function of the pediatric oncology nurse is to provide symptom management to children with cancer. Symptom management strategies have been published, but there is scarce literature examining neither the actual use of these nursing interventions, nor the effects of using these interventions on the nurses' perceived work environment. The purpose of this study was to examine the nursing interventions used in treating pediatric oncology patients' symptoms, as well as the emotional sequelae from providing this care. Phase One of this study examined the content validity of the newly developed Nurses Distress and Interventions for Symptoms Survey (NDISS) utilizing content experts. Phase Two of this study involved both the reliability testing of the NDISS by test-retest and served as a pilot for Phase Three.^ In Phase Three, a national sample of pediatric oncology nurses was surveyed about their patients' symptoms, the nurses' distress from the symptoms, the nursing interventions used to treat the symptoms, the perceived efficacy of the nursing interventions, and their job satisfaction. The response rate was 53%, and analysis of study hypotheses were evaluated using Pearson's correlation and multiple regression analyses. The main study variables were not related in the hypothesized direction; therefore four of the six hypotheses were not supported. However, quantity and perceived effectiveness of nursing interventions were both found to act as mediators in the study model, and as a result, these two hypotheses were retained.^ The results of the survey demonstrated a high frequency of distressing patient symptoms as perceived by nurses (mean 6, range 0-7); nurses rated their distress from these symptoms as moderate (mean 2.9, range 0-4); nurses used an average of 12.7 nursing interventions per symptom (range 0-38); nurses found the nursing interventions moderately effective (mean 2.5, range 0-4); and nurses had moderately high overall job satisfaction (mean 3.9, range 1-5). Although many of the hypotheses were not supported, interesting trends in the data were found. In addition, the findings provided elucidation of specific nursing interventions used by pediatric oncology nurses as well as a description of the effects of providing patients' symptom management, including nurses' distress, perceived effectiveness of nursing interventions, and job satisfaction.
502
Dissertation (Ph.D.)--University of South Florida, 2007.
504
Includes bibliographical references.
516
Text (Electronic dissertation) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
500
Title from PDF of title page.
Document formatted into pages; contains 90 pages.
Includes vita.
590
Adviser: Susan C. McMillan, Ph.D.
653
Children.
Cancer.
Distress.
Job satisfaction.
Nursing interventions.
690
Dissertations, Academic
z USF
x Nursing
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.2199



PAGE 1

Relationship Between Nu rses’ Management of Pediatric Oncology Patients’ Symptoms and Job Satisfaction by Jennifer I. Rheingans A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy College of Nursing University of South Florida Major Professor: Susan C. McMillan, Ph.D., A.R.N.P. Lois Gonzalez, Ph.D., A.R.N.P. Janine Overcash, Ph.D., A.R.N.P. Brent J. Small, Ph.D. Date of Approval: October 22, 2007 Keywords: children, cancer, distress, job satisfaction, nursing interventions Copyright 2007, Jennifer I. Rheingans

PAGE 2

Acknowledgements This work absolutely could not have been done without the professionalism, insight, generosity, suppor t, and time from hundreds of national pediatric oncology nurses. The author gratefully acknowle dges the American Cancer Society for the financial support fr om a doctoral scholarship. Finally, the author sincerely appreciates the patienc e, kindness, and support from her dissertation committee, family and fr iends over the past four years.

PAGE 3

i Table of Contents List of Tables iii List of Figures iv Abstract v Chapter One – Introduction 1 Problem Statement 4 Purpose 5 Research Questions 5 Definition of Terms 6 Job Satisfaction 6 Job-related Stressors 6 Nurses’ Distress 6 Nursing Interventions 6 Nurses’ Perceived Effectiveness of Nursing Interventions 7 Quantity of Nursing Interventions 7 Pediatric Oncology 7 Pediatric Oncology Nursing 7 Symptom Management 7 Symptom Distress 8 Significance of the Study 8 Chapter Two – Review of Literature 9 Conceptual Model 9 Pediatric Oncology Patients’ Symptoms 12 Nurses’ Distress 13 Nursing Interventions 14 Job Satisfaction 16 Chapter Summary 17

PAGE 4

ii Chapter Three – Methods 19 Sample 20 Phase One 20 Phase Two 20 Phase Three 21 Instruments 22 Nurses’ Distress and Interventions for Symptoms Survey 22 Measure of Job Satisfaction 24 Demographics 25 Procedures 26 Institutional Approvals 26 Phase One 26 Phase Two 27 Phase Three 28 Data Analysis 29 Phase One 29 Phase Two 29 Phase Three 30 Chapter Four – Results 32 Phase One 32 Phase Two 32 Phase Three 37 Sample Description 37 Bivariate Correlations 43 Regression Analyses 44 Chapter Five – Discussion 51 Phase One 51 Phase Two 52 Phase Three 55 Sample Description 55 Bivariate Correlations 63 Regression Analyses 66 Conclusions 68 References 71 Appendices 81 Appendix A: Nurses’ Distress and Interventions for Symptoms Survey 82 Appendix B: Nurses’ Ranked Use of Interventions in Phase Three 89 About the Author End Page

PAGE 5

iii List of Tables Table 1 Summary of Phase Two Nurses’ Age and Years in Practice. 34 Table 2 Summary of Phase Two Demographics. 35 Table 3 Summary of Phase Two Nursing Practice Characteristics. 36 Table 4 Summary of Phase Three Nurses’ Age and Years in Practice. 38 Table 5 Summary of Phase Three Demographics. 39 Table 6 Summary of Phase Three Nursing Practice Characteristics. 40 Table 7 NDISS Phase Three Summary Results of Symptoms Present. 41 Table 8 NDISS Phase Three Summary Results. 41 Table 9 Measure of Job Satisfaction (MJS) Results for Phase Three. 43 Table 10 Correlations Among Main Study Variables. 45 Table 11 Correlation Table of Depende nt Variables and Covariates. 46 Table 12 Regression Findings Evaluating Perceived Effectiveness. 48 Table 13 Regression Findings Evaluating Quantity of Nursing Interventions. 49 Table 14 Regression Findings Evaluating Study Model on Job Satisfaction. 50

PAGE 6

iv List of Figures Figure 1 The Stress Response Seque nce Model (Hinds et al., 1990). 10 Figure 2 The Stress Response Sequence Model in Pediatric Oncology Nursing Symptom Management. Adapted from Hinds, et al. (1990). 11

PAGE 7

v Pediatric Oncology Nurses’ M anagement of Patients’ Symptoms Jennifer I. Rheingans ABSTRACT A primary function of the pediatric onc ology nurse is to provide symptom management to children with cancer. Sy mptom management strategies have been published, but there is sca rce literature exam ining neither the actual use of these nursing interventions, nor the effect s of using these interventions on the nurses’ perceived work environment. The pur pose of this study was to examine the nursing interventions used in treating pediatric oncology patients’ symptoms, as well as the emotional sequelae from providing this care. Phase One of this study examined the content validity of the newly developed Nurses Distress and Interventio ns for Symptoms Survey (NDISS) utilizing content experts. Phase Two of this study involved both the reliability testing of the NDISS by te st-retest and served as a pilot for Phase Three. In Phase Three, a national sample of pediatric oncology nurses was surveyed about their patients’ symptom s, the nurses’ distress from the symptoms, the nursing inte rventions used to treat t he symptoms, the perceived efficacy of the nursing interv entions, and their job satisfaction. The response rate was 53%, and analysis of study hypotheses were evaluated using Pearson’s correlation and multiple regression analyses.

PAGE 8

vi The main study variables were not re lated in the hypothesized direction; therefore four of the si x hypotheses were not suppor ted. However, quantity and perceived effectiveness of nursing interv entions were both found to act as mediators in the study model, and as a result, these two hypotheses were retained. The results of the survey dem onstrated a high frequency of distressing patient symptoms as perceived by nurses (mean 6, range 0-7); nurses rated their distress from these symptoms as moder ate (mean 2.9, range 0-4); nurses used an average of 12.7 nursing interventi ons per symptom (range 0-38); nurses found the nursing interventions moderatel y effective (mean 2.5, range 0-4); and nurses had moderately high overall job satisfaction (mean 3.9, range 1-5). Although many of the hypotheses were not supported, interesting trends in the data were found. In addition, the findi ngs provided elucidation of specific nursing interventions used by pediatric oncology nurses as well as a description of the effects of providing patients’ symptom management, including nurses’ distress, perceived effectiveness of nursi ng interventions, and job satisfaction.

PAGE 9

1 Chapter I Introduction Nearly 10,000 children between birt h and 14 years old are diagnosed with cancer each year, making pediatric c ancer the number one disease killer of children (American Cancer Society, 2006; CureSearch, n.d.). Despite the tragedy of pediatric oncology, there is great hope – overall pediatric c ancer survival was estimated to be 79% in 2006 (American Cancer Society, 2006). Such great success in the improvement of survival rates (up from 50% in the 1970’s) is obtained at the expense of intense treatm ent protocols, oft en inducing a breadth of distressing symptoms and long-term negative effects from the cancer treatment. In fact, pediatric oncology pati ents continue to rate symptoms as the overall most difficult aspect of cancer treatment (Hedstrm, Haglund, Skolin, & von Essen, 2003; Moody, Meyer, M ancuso, Charlson, & Robbins, 2006; Woodgate, 2005, 2006; Woodgate & Degner 2004; Woodgate, Degner, & Yanofsky, 2003). Patients’ symptoms further challe nge pediatric oncology as the nurses struggle to treat symptoms. Patients’ symptom distress has been documented as a primary contributor to the highly stre ssful nursing environment in pediatric and adult oncology settings (Barnard, Street, & Love, 2006; Bond, 1994; de Carvalho, Muller, de Carvalho, & de Souza Melo, 2005; Emery, 1993; Hinds et al., 2003;

PAGE 10

2 Kushnir, Rabin, & Azulai, 1997; Papadatou, Bellali, Papazoglou, & Petraki, 2002; Petrova, Todorova, & Mateva, 2005). Inte rnationally, oncology nurses have rated symptom management among the most important research priorities, including nurses in Canada (Fitch, Bakker, & C onlon, 1999), the United Kingdom (Soanes, Gibson, Bayliss, & Hannan, 2000; Soanes Gibson, Hannan, & Bayliss, 2003), and the U.S. (Cohen, Harle, Woll, Despa, & Munsell, 2004). Thus, unrelieved symptoms remain a significant problem for both patients and nurses in pediatric oncology. Traditional medical-based nursing care emphasizes the use of pharmacologic agents to address patients’ symptoms (Panzarella et al., 2002). Pharmacologic symptom management has in fact made great strides. However, after pharmacologic symptom management strategies have been exhausted, and, despite efforts to provide holistic patient care, nursing interventions to manage symptoms become less clearly defined. There is no discrete recipe for symptom management in pediatric oncology. Recommendations abound for potentially useful nursing interventions, describing both pharmacologic and nonphar macologic options (Baggott, Kelly, Fochtman, & Foley, 2002; Ladas, Post -White, Hawks, & Taromina, 2006; National Comprehensive Cancer Networ k, 2007d). However, there is little information available describing which nur sing interventions are available or actually used by nurses in general or, mo re specifically, in oncology patient populations. The few published studies found measured nursing interventions

PAGE 11

3 among all nursing popul ations (Bulecheck, McCloske y, Titler, & Denehey, 1994) or measured the use of only select nur sing interventions (e.g., nonpharmacologic or complementary and alternative ther apies) (Helmrich et al., 2001; Hessig, Arcand, & Frost, 2004; King, Pettigrew, & Reed, 1999; Rankin-Box, 1997; Tracy et al., 2005). A description of nurses’ in terventions are needed provide insight into the nurses’ experience of distress rela ted to patients’ sympt oms. In order to design future clinical outcome studies, it is important to doc ument which nursing interventions are currently being used. Although treating patients’ symptoms is a critical issue, research literature supports the fact that nurse s tend to remain in a stressful environment only when they feel satisfied with their work. On cology nurses (both pediatric and adult) particularly enjoy the richness and reward in making relationships with patients and families (Bertero, 1999; Clarke-Steffen, 1998; Cohen, Haberman, Steeves, & Deatrick, 1994; Cohen & Sa rter, 1992; Fall-Dickson & Ro se, 1999; Grunfeld et al., 2005; Haberman, Germino, Maliski, Stafford-Fox, & Ric e, 1994; Olson et al., 1998; Papadatou et al., 2002), as well as fee ling the comfort of knowing that they have improved the lives of their patients and families through their nursing care (Clarke-Steffen, 1998; C ohen et al., 1994; Fall-Dickson & Rose, 1999; Haberman et al., 1994; Olson et al., 1998; Papadatou et al., 2002; Papadatou, Martinson, & Chung, 2001). This information suggests t hat employers and researchers should examine the conditions whic h support job satisfaction and retain nurses despite the stressful nursing environm ent of pediatric oncology.

PAGE 12

4 Problem Statement Pediatric oncology patients continue to suffer from cancer diseaseand treatment-related symptoms. This suffering is one of the primary causes of jobrelated stress for pediatric oncology nurses. Nurses feel particularly frustrated when patients continue to su ffer and they, as care pr oviders, have no further interventions to offer. In th is day of safety priorities and nursing shortages, it is important to focus on both the care of the patients as well as the working conditions of the nurses. An opportunity ex ists to study the potential relationship between nurses’ symptom managem ent and job satisfaction. There is a lack of published research in which the symptom management process is examined from the nurses’ perspective, particularly in pediatric oncology nursing. Specifically missing fr om the literature is data on nurses’ appraisal of patients’ symptom s, nursing interventions used to help treat patients’ symptoms, and nurses’ subseque nt level of job satisfaction. It is posited that nurses who are satisfied with their effe ctiveness in treating patients’ symptoms will experience less distress t han those who feel less able to alleviate distressing symptoms. The use of a lar ger repertoire of nursing interventions may also help to protect the nurse from feeling s he/he has exhausted all symptom management possibilities and must therefore watch the patient suffer without hope. Job satisfaction is seen as an outcome of nursing practice and has been noted to reflect the nurses’ perception of her care Therefore, degree of job satisfaction

PAGE 13

5 may be related to the degree of nurses’ di stress from less than expected patient symptom management. Purpose The purpose of this descriptive corre lational study was to describe the symptom management experience of pediatric oncology nurses and the relationship between nurses’ symptom management and job satisfaction. The variables examined included the presence of patients’ distressing symptoms as perceived by the nurse, nurses’ distre ss from these symptoms, the nursing interventions used to treat these symptoms, the perce ived effectiveness of these nursing interventions, and nurses’ job satisfaction (see Figure 2, page 13). Research Questions The following hypotheses guided the study: 1. There is a positive relationshi p between the presence of distressing symptoms in pediatric oncology patients and the nurses’ distress from those symptoms. 2. There is an inverse relationship bet ween the nurses’ perceived effectiveness in treating patients’ symptoms and t he nurses’ distress from patients’ symptoms. 3. There is an inverse relationship bet ween the number of nur sing interventions used to treat these symptoms and t he nurses’ distress from patients’ symptoms.

PAGE 14

6 4. There is an inverse relationship between nurses’ distress and nurses’ job satisfaction. 5. The nurses’ perceived effectiveness of nursing interventions acts as a mediator between patients’ sym ptoms and nurses’ distress. 6. The quantity of nursing interventions ac ts as a mediator between patients’ symptoms and nurses’ distress. Definition of Terms Job Satisfaction The extent of positive affective ori entation to the job (Traynor & Wade, 1993). Job-related Stressors “Activators or determinants of the condition, including internal and external environmental events or condition s that change an individual’s present state and may produce notabl e physical or psychosocial reactions” (Hinds, Quargnenti, Hickey, & Mangum, 1994, p. 62) Nurses’ Distress The amount of distress experienced by pediatric oncology nurses related to witnessing patients who are ex periencing distressing symptoms. Nursing Interventions Defined by Nursing Interventions Classif ication (NIC) as “any direct care treatment that a nurse performs on behal f of a client. The treatments include nurse-initiated treatments resulting from nursing diagnoses, physician-initiated

PAGE 15

7 treatment resulting from medical diagn oses, and performance of daily essential functions for the client who cannot do these.” (Bulecheck & McCloskey, 1999, p. 23) Nurses’ Perceived Effectiveness of Nursing Interventions The nurses’ perception of the effectiv eness of her/his nursing interventions to treat patients’ symptoms. Quantity of Nursing Interventions The number of nursing interventions used by a nurse to treat patients’ symptoms. Pediatric Oncology The healthcare management of child ren with cancer – aged infancy through adolescence (Baggott et al., 2002). Pediatric Oncology Nursing The practice of the subspecialty of nursing in the field of pediatric oncology (Foley & Ferguson, 2002); as ident ified by membership in the national Association of Pediatric Hematol ogy Oncology Nurses (APHON). Symptom Management The process of treating diseaseand treatment-related complications, including multidisciplinary approaches, such as surgical, pharmacologic, or nursing.

PAGE 16

8 Symptom Distress “The degree or amount of physical or mental upset, anguish, or suffering experienced with a specific sympto m” (Rhodes & Watson, 1987, p. 243). Significance of the Study Studies are needed in order to build on prior research calling for further attention to pediatric oncology patient s’ symptom management as well as address the gaps in knowledge surroundi ng nurses’ symptom management. This study was based on literature describing high levels of nursing stress and distress in managing patients’ symptoms. Attention to nursing outcomes may help to focus interventions for recruiting and retaining quality pediatric oncology nurses in a highly-stressful work environment. Given the paucity of data regarding pediatric oncology nurses’ symptom management strategies, the measurement of those currently used nursing interventions will help to establish basel ine data. This may lead to clinical interventional studies desig ned to improve comfort and treatment for the child and family. Future studies may build on this data by investigating causal relationships between specific nursing inte rventions and the resultant patient and nursing outcomes (e.g., symptom distre ss or nurses’ job satisfaction, respectively).

PAGE 17

9 Chapter II Review of Literature This chapter presents the review of literature. Fi rst the conceptual framework is delineated. Then empirical st udies are synthesized according to the main variables in this study, including pediatric oncology patients’ symptoms, nurses’ distress from patients’ symptoms, nurses’ use of nursing interventions to manage patients’ symptoms, and nurses’ j ob satisfaction. Finally, a summary of the relationships among thes e variables is provided. Conceptual Model The Stress Response Sequence Model (SRSM) was designed in 1982 by a study group consisting of Institute of Medicine and the National Academy of Sciences (Elliott & Eisdorfer, 1982; Hi nds et al., 1998). The SRMS was based on the review of research and models of the stress process in humans and since has been used extensively to guide the study of the stress process in pediatric oncology nursing (Clarke-Steffen, 1998; Hi nds, 2000; Hinds et al., 1990; Hinds et al., 2000; Hinds et al., 1994; Hinds et al., 1998; Hinds et al., 2003; Olson et al., 1998). The SRSM consists of four co re concepts: stressors, reactions, consequences, and mediators (see Figure 1) In the SRSM, the stressors are internal or external and are defined as environmental events or conditions that

PAGE 18

10 impact or alter an individual’s current st ate” (Hinds et al., 1990). Reactions are the biological or psychological respons es to the stressors. Consequences are a result of reactions, are classified in the categories of biological, psychological, or sociological, and can be evaluated as pos itive or negative. Mediators in the SRSM are filters and modifiers that a ffect the stressor-reaction-consequence sequence and may cause individual variati ons. The SRSM emphasizes the fluid interplay among all of the concepts. Stressors Mediators Reactions Consequences Figure 1. The Stress Response Sequenc e Model (Hinds et al., 1990)

PAGE 19

11 Figure 2 demonstrates the adaptation of the SRSM for this study. The presence of patients’ symptoms is the init ial stressor in this model. The nurses’ reaction in this model is the amount of distress the nurse feels as a result of patients’ distressing symptom s (e.g. her/his appraisal of the patients’ distress). Job satisfaction is conceptualized as the consequence in this model. Nurses’ Perceived Effectiveness of Nursing Interventions Nurses’ Distress from Patients’ Symptoms Number of Patients’ Symptoms Present Nurses’Job Satisfaction Quantity of Nursing Interventions (-) (+) (-) (-) Stressor Mediators Reaction Consequence Figure 2. The Stress Response Sequence M odel in Pediatric Oncology Nursing Symptom Management. Adapted from Hinds, et al. (1990).

PAGE 20

12 The mediators in the current study are conceptualized as affecting the relationship between the stressor (patient s’ symptoms) and the reaction (nurses’ distress). There are two proposed mediat ors: the perceived effectiveness of the nursing interventions and the quantity of the nursing interventions. Nursing interventions are hypothesized to affect the way nurses interpret patients’ symptoms by offering a mechanism fo r treating patient s’ symptoms. Pediatric Oncology Patients’ Symptoms Pediatric oncology patients rate symp toms as the overall most difficult aspect of cancer treatment (Hedstrm et al., 2003; Moody et al., 2006; Woodgate, 2005, 2006; Woodgate & Degner 2004; Woodgate et al., 2003). Symptoms are defined as a patient’s s ubjective description or expression of a disease or a change in condition (Ander son, Anderson, & Glanze, 1994). The distress or bother caused from patient s’ symptoms has increasingly been the focus of both adult and pediatric resear ch (McMillan & Small, 2002; Rhodes & Watson, 1987; Woodgate, 2005, 2006). Sympto m distress is a more subjective personal interpretation of the effect of having the symptoms – as compared to frequency (how often the symptoms occur) or severity (how much of the symptom is present) as symptoms had been traditionally measured (Rhodes & Watson, 1987). According to pediatric onc ology patients, the most distressing symptoms are (in descending orde r) fatigue, pain, poor appetite, nausea/vomiting, hair loss, isolation, wo rry, fear, mouth so res, trouble with movement, trouble with relationships, and trouble sleeping (Collins et al., 2002;

PAGE 21

13 Drake, Frost, & Collins, 2003; Enskar, Carlsson, Hamrin, & Kreuger, 1996; Hedstrm et al., 2003; Hicks, Bartholom ew, Ward-Smith, & Hutto, 2003; Hinds, Quargnenti, & Wentz, 1992; Jalmsell, Kreicbergs, On elov, Steineck, & Henter, 2006; McCaffrey, 2006; Moody et al., 2006; Novakovic et al., 1996; Wolfe et al., 2000). Nurses’ Distress The detrimental effects of patients’ symptoms are noted in nurses as well. Patients’ symptom distress is a primary c ontributor to the hi ghly stressful nursing environment of pediatric (and adult) oncol ogy (Barnard et al., 2006; Bond, 1994; de Carvalho et al., 2005; Emery, 1993; Hin ds et al., 2003; Kushnir et al., 1997; Papadatou et al., 2002; Petrova et al., 2005). Nurses find pleasure and meaning in helping children and families (ClarkeSteffen, 1998; Cohen et al., 1994; FallDickson & Rose, 1999; Haberman et al., 1994; Olson et al., 1998; Papadatou et al., 2002; Papadatou et al., 2001); however unrelieved symptoms cause nurses to feel highly stressed and anxious bec ause they feel they have no further interventions to help treat the patients’ suffering (Barnard et al., 2006; Bond, 1994; Clarke-Steffen, 1998; Cohen et al., 1994; C ohen & Sarter, 1992; de Carvalho et al., 2005; Ergun, Oran, & B ender, 2005; Fall-Dickson & Rose, 1999; Florio, Donnelly, & Zevon, 1998; Kus hnir et al., 1997; Olson et al., 1998; Papadatou et al., 2002; Papadatou et al., 2001; Petrova et al., 2005). An example of a study examining this e ffect was conducted in the adult oncology nursing environment (McMill an et al., 2006). Researchers adapted the Memorial

PAGE 22

14 Symptom Assessment Scale (MSAS) – a valid and reliable tool used to measure patients’ symptom distress (Portenoy et al., 1994) – to reflect the caregiver’s distress as a result of the patient’s symptoms. Nursing Interventions Nursing interventions are treatm ents based on clinical judgment and knowledge used by nurses to enhanc e patient outcomes (Bulecheck & McCloskey, 1999). These interventions invo lve both direct and indirect care; and are initiated by nurses, physicians, and other healthcare providers. A primary function of pediatric oncology nursing is the provision of symptom management using nursing interventions (Associati on of Pediatric Hematology/Oncology Nurses, 2007). Nursing care of symptom managem ent has traditionally emphasized medical-based administrati on of pharmacologic agents (Panz arella et al., 2002). Pharmacologic management has in fact made great strides. For example, the development of 5-HT3 blockers for nausea offer improved management of chemotherapy-induced nausea and vomiting, without many of the side effects seen in previous anti-emetics (e.g. s edation, extrapyramidal effects, etc.) (Antonarakis et al., 2004). Yet despite t hese medications, pat ients continue to complain of distress from nausea/vomiti ng (Collins et al., 2000; Collins et al., 2002; Hedstrm et al., 2003; Moody et al., 2006; Novakovic et al., 1996). More recent texts have started recommending a more holistic approach to symptom management, including both pharmacologi c and nonpharmacologic interventions

PAGE 23

15 (Baggott et al., 2002; Dossey, Keegan, & Guzzetta, 2005; National Comprehensive Cancer Networ k, 2007a, 2007b, 2007c, 2007d; Wong, Hockenberry, Wilson, Winkelstein, & Kline, 2003). Examples of some of the more commonly recommended nonpharmacologic in terventions include acupuncture, acupressure, art therapy, deep breathing, distraction, humor, imagery, massage, music therapy, and pet therapy. Although there are now recommendati ons for nursing interventions in symptom management, there is little published research describing which nursing interventions are in fact bei ng used by nurses for symptom management. There is also little data describing the effectiveness of many of the recommended interventions. A description of nurses’ interventions, including perceived effectiveness of these interventions, ma y provide insight into the nurses’ experience of managing patients’ sympt oms. This information may help in examining nursing interventions as t hey related to nurses’ distress. A search for published research on nursing interventions for symptom management led only to a dated article of the frequency of general nursing interventions (Bulecheck et al., 1994) and a handful of surveys specifically assessing nurses’ use of complementary an d alternative therapies (Helmrich et al., 2001; Hessig et al., 2004; King et al., 1999; Rankin-Box, 1997; Tracy et al., 2005). While there is no data in pediatric oncology, nurses in adult populations (including oncology, critical care, and gener al nursing) report t he frequent use of diet, exercise, massage, prayer/spiritual ity, relaxation, and visualization for

PAGE 24

16 patient care (Helmrich et al., 2001; Hessi g et al., 2004; King et al., 1999; RankinBox, 1997; Tracy et al., 2005). Thes e surveys were intended to measure complementary and alternative therapies specifically and therefore did not measure a comprehensive list of nursing interventions. There was no data to describe which nursing interventions were used specifically for symptom management, including in pediatric oncology. Job Satisfaction Job satisfaction has been extensively studi ed in the research literature in attempts to retain nurses in the stressful healthcare environment (Blegen, 1993). In a recent survey of oncology nursing research priorities, oncology nurses rated job satisfaction as the third most impor tant out of 120 choices (Cohen et al., 2004). Job satisfaction describes a person’s positive affective appraisal of one’s job (Traynor & Wade, 1993, p. 127; Wade, 1993). This is consistent with the SRSM as job satisfaction has been prev iously conceptualized as a consequence of pediatric oncology nursing (Hinds et al., 1998; Hinds et al., 2003). Pediatric and adult oncology nurses par ticularly enjoy the richness and reward in making relationships with pat ients and families (Bertero, 1999; ClarkeSteffen, 1998; Cohen et al., 1994; Cohen & Sarter, 1992; Fall-Dickson & Rose, 1999; Grunfeld et al., 2005; Haberman et al., 1994; Olson et al., 1998; Papadatou et al., 2002). Oncology nurses ta ke great comfort knowing that they have improved the lives of their patients and families through their nursing care

PAGE 25

17 (Clarke-Steffen, 1998; C ohen et al., 1994; Fall-Dickson & Rose, 1999; Haberman et al., 1994; Olson et al., 1998; Papadatou et al., 2002; Papadatou et al., 2001). Job satisfaction in pediatric oncology nursing has been positively correlated with coworker cohesiveness and organizational commitment, and negatively correlated with a nurses intent to leave an organization (Hinds et al., 1998; Steen, Burghen, Hinds, Srivastava, & Tong, 2003) and job stress (Hinds et al., 1998). No literature was found whic h examined the potential relationship between job satisfaction as a result of nursing interventions. Chapter Summary This study is based on the St ress Response Sequence Model and examines the relationship between t he following variables: nurses’ perceived presence of distressing symptoms in pediat ric oncology patients; nurses’ distress from these symptoms; a descr iption of the nursing interventions used to manage these symptoms; nurses’ perceived effect iveness of these interventions; and the job satisfaction of the pediatri c oncology nurses. Nurses feel distress as a result of patients’ distress from symptoms. Nurses attempt to manage patient symptoms with nursing interventions. The perceived degree of success in managing these symptoms may affect the nurses’ distress. Job satisfaction among pediatric oncology nurses has been docum ented as directly related to the quality and meaning of care provided to the children. Therefore, nurses’ distress may affect nurses’ perceived work environm ent, specifically job satisfaction. The purpose of this study is to examine which nursing interventions are being used by

PAGE 26

18 pediatric oncology nurses to manage patie nts’ most distressing symptoms, and to evaluate these interventions as potentia l mediators in the stress process of symptom management for pedi atric oncology nurses.

PAGE 27

19 Chapter III Methods Chapter Three presents study methods This three-phase study consisted of psychometric data gathering for a new instrument and a randomized crosssectional descriptive correlational survey using the newly developed instrument. As there was little in the research liter ature examining the pr oposed relationships of the variables in this study, a descriptive approach was considered to be the most appropriate as a foundat ional step in examining ne w conceptual linkages or hypotheses (Polit & Beck, 2004). Phase One examined the content validity of the Nurses’ Distress and Interventions for Symptoms Survey ( NDISS). Phase Two was the reliability testing of the NDISS using test-retest, as well as the piloting of the multiple contacts method and questionnaire packet for Phase Three. Phase Three was a national survey of pediatric oncology nurses examining nurse s’ management of patients’ symptoms and job satisfaction. This chapter will describe the study by phase and conclude with a brief summary.

PAGE 28

20 Sample Phase One In Phase One, no subjects were approached. Content experts advised the investigators about which nursing intervent ions to include in the questionnaire. Twenty pediatric oncology nurses were identified through two APHON Listservs to serve as expert consultants for the first part of Phase One. For the second part of Phase One, seven expert consultants were selected to examine the cont ent validity of the newly de veloped NDISS, including five APHON Listserv respondents and two professional pediatric oncology nurse researchers. Five of the par ticipants were certified in pediatric oncology nursing. Phase Two A sample of 100 pediatric oncology nurses was sought from the local chapter of the Association of Pediatri c Hematology/Oncology Nurses (APHON), APHON Listservs, and the national APHON membership list (see Sample: Phase Three for details). Inclusion criteria fo r the sample were i dentification of the nurses as registered nurse (RN) s pecializing in pediatric oncology as demonstrated by APHON membership, with a minimum of six months of pediatric oncology nursing experience; the ability to read and write in English as demonstrated by the completion of the survey packet; and direct-patient-care practice as indicated on the demographics sheet.

PAGE 29

21 Phase Three 1,000 pediatric oncology nurses were solicited for participation in Phase Three of this study. APHON is a national professional pediatric oncology nursing organization with approximately 2,500 inter national members. A membership list of 1,200 names and physical addresses was purchased from APHON. The random sample was derived by APHON using systematic random sampling and was limited to direct-patient care nur ses residing in the United States. Any participants from Phase Two were remov ed from the Phase Three mailing list. A conservative estimate of mail-based su rvey response rates suggested that 1025% will respond (Dillman, 2007). Based on a two-tailed correlational analysis with a small-medium correlation (r=.20), a sample size of 191 participants was required to achieve 80% power (Faul & Erdfelder, 1992). Inclusion criteria for the sample we re identification of the nurses as registered nurse (RN) specializing in pediatric oncology as demonstrated by APHON membership, with a minimum of six months of pediatric oncology nursing experience; the ability to read and write in English as demonstrated by the completion of the survey packet; direct -patient-care practice as indicated on the demographics sheet; and a practice site in the U.S. as identified in the APHON membership address roster.

PAGE 30

22 Instruments Nurses’ Distress and Interventions for Symptoms Survey (NDISS) The NDISS is provided in the Appendi x and is available for unrestricted use. The NDISS was created for this resear ch project after an extensive review of the literature revealed no relevant su rveys previously designed. The purpose of the NDISS was twofold: 1) to assess the nurses’ appraisal of patients’ symptoms and symptom distress; and 2) to assess the use of nursing interventions in treating patients’ sym ptoms. The symptoms included in the NDISS were based on a literature review of the most distressing symptoms according to children with cancer and include fatigue, pain, poor appetite, nausea/vomiting, hair loss, isolation, wo rry, mouth sores, and trouble sleeping. Validity testing for content was conducted by using content experts to determine a content validity index (see Procedure: Phase One). Reliability testing for testretest stability was conducted in Phas e Two by the completion of two NDISS surveys, sent out two weeks apart (see Procedure: Phase Two). For each of the seven sym ptoms, participants were asked about relevance [“In the past month, have any of y our patients experienced (symptom)?”]. Frequency counts were made per symptom across participants, and a mean was assessed of the number of symptoms repor ted as present by each nurse. If the participant responded affirmatively about the presence of a symptom, then the participant was asked to rate her or his own distress from t he patients’ symptom using a 5-point summated rating scale (from “0 not at all” to “4 very much”).

PAGE 31

23 This was based on the Memorial Sympto m Assessment Scale (MSAS) (Collins et al., 2000; Collins et al., 2002). McMillan, et al. (2006) used a similar approach by assessing caregiver’s distress from patient s’ symptoms in a modified version of the Memorial Symptom Assessment Scale, and found the Caregiver version to be valid and reliable. The nurses’ di stress scores on the NDISS were averaged per symptom across participants and mean distress scores were calculated by averaging each nurses’ distress sco re for all applicable symptoms. Following the distress question, the par ticipant was asked about her/his own use of nursing interventions. They were offered a list of nursing interventions and asked to choose which of the nursing interventions they “normally” use to treat that particular sympt om. The list of nursing interventions was the same for all symptoms and included space for the writ e-in of additional interventions not listed. The nursing interventions were derived from a pediatric nursing reference text (Wong et al., 2003), a pediatric oncol ogy nursing text (Baggott et al., 2002), a holistic nursing text (Dossey et al., 2005), an article on the Nursing Interventions Classifications (NIC) Projec t (Bulecheck et al., 1994), and the most current evidence-based practice guidel ines published by a leading national oncology advocacy organization (National Comprehensive Cancer Network, 2007a, 2007b, 2007c, 2007d). The list of 64 pot ential interventions was reduced to 35 interventions as a result of a surv ey of pediatric oncology nurse experts for content validity (see Procedure: Phase O ne). To score this section, the number of nursing interventions utilized per symptom was summed. The number of

PAGE 32

24 interventions used was averaged across sy mptoms resulting in a “number of nursing interventions” score. This was also referred to as the quantity of nursing interventions. The final NDISS question for each symptom asked the nurse to rate her/his perceived effectiveness in treati ng each symptom using the interventions chosen from the list. Participants res ponded by using a 5-point summated rating scale (from “0 not at all” to “4 ve ry much”). The nurses’ effectiveness scores on the NDISS were averaged resulting in a “nurses’ perceived effectiveness” score. Measure of Job Satisfaction (MJS) The MJS has been used extensively in the testing of the Stress Response Sequence Model (Hinds et al., 1998; Hinds et al., 2003). The MJS was designed to measure nurses’ “positive affective ori entation to their job” (Traynor & Wade, 1993, p. 128) and contained 43-items in seven subscales with Cronbach’s alpha ranging from 0.85-0.90: Personal Satisf action, Satisfaction with Workload, Satisfaction with Professional Support, Sati sfaction with Training, Satisfaction with Pay, Satisfaction with Prospects, and Satisfaction with Standards of Care (Traynor & Wade, 1993; Wade, 1993). Each question used a summated rating scale from 1 (very dissatisfi ed) to 5 (very satisfied). Qu estions for each subscale were averaged, yielding a subscale sco re (range 1-5), with a higher score indicating greater job sati sfaction. An overall satisfaction score was also calculated which represented an aver age of all 43 items (range 1-5). Job

PAGE 33

25 satisfaction was conceptualized in this st udy as the overall job satisfaction score, though subscales were also examined to provide insight. Demographics A demographics form was included to measure background nursing information, including nurses’ de mographics (i.e., age, gender, geographic location, religious background, ethnicity and level of education), and the nurses’ practice characteristics (i.e., current prac tice role, practice setting, type of nursing position, years in nursing, years in pediatric nursing, presence of pediatric oncology nursing certification, and inst itutional Magnet st atus). Geographic location was categorized according to the divisions used by the U.S. Census Bureau: New England (Connecticut, Mai ne, Massachusetts, New Hampshire, Rhode Island and Vermont); Middle Atlantic (New Jersey, New York, and Pennsylvania); East North Central (I llinois, Indiana, Michigan, Ohio, and Wisconsin); West North Central (Iowa, Kansas, Minnesota, Missouri, Nebraska, North Dakota and South Dakota); South Atla ntic (Delaware, District of Columbia, Florida, Georgia, Maryland, North Caro lina, South Carolina, Virginia, and West Virginia); East South Central (Alabama, Kentucky, Mississippi and Tennessee); West South Central (Arkansas, Loui siana, Oklahoma and Texas); Mountain (Arizona, Colorado, Idaho, Montana, Nev ada, New Mexico, Utah and Wyoming); and Pacific (Alaska, California, Hawaii, Oregon, and Washington) (U.S. Census Bureau, 2007). There were no requests fo r information which would personally identify a participant.

PAGE 34

26 Procedures Institutional Approvals IRB approval was obtained by the University of South Florida’s Institutional Review Boar d. Expedited review was gr anted as there was no linking of personal identification information to the surveys; and an exemption was also granted to allow the return of the survey to suffice as informed consent approval. Permission was obtained through t he Association of Pediatric Hematology/Oncology Nurses (APHON) prior to IRB application in order to solicit membership for st udy participation. Phase One Phase One was the content validity test ing of the NDISS. The first part of Phase One used pediatric oncology nurse experts to analyze the choices of nursing interventions in the NDISS. Respon dents were sent a list of 64 nursing interventions recommended in the literatur e for treating the seven symptoms on the NDISS. The nurses were asked to sele ct the nursing interventions “normally” used to treat the symptoms. Based on t he responses, the most frequently used interventions were retained for further versions of the NDISS. The second part of Phase One involved the content validity testing of the NDISS using a voluntary panel of exper t consultants in pediatric oncology nursing. Participants were sent the NDISS, the resource reference list used to design the NDISS, an evaluation form, a $5 Starbuck’s gift card as a token of appreciation, and a pre-stamped return env elope. The evaluation form asked the

PAGE 35

27 participants to determine whether each item matched the survey objectives (response choices were “yes” with a score of 1, “no” with a score of -1, or “uncertain” with a score of 0) (McMillan, 1990; McMill an, Williams, Chatfield, & Camp, 1988). The result of the evaluatio ns was a content validity index (CVI) (see Analysis: Phase One) which was us ed to assess the validity of the newly developed NDISS. Phase Two The purpose of Phase Two was to demons trate the reliability of the NDISS using a test-retest method, as well as to pilot test for Phase Three. The Dillman Tailored Design Method (Dillman, 2007) advocated for multiple contacts, therefore in the first mailing series of Phase two, participants were sent a prenotification letter (day -3), followed by a questionnaire packet (day 0), and a follow-up thank you/reminder postcar d (day 7). The questionnaire packet contained an IRB-approved introductory le tter serving as informed consent, a numbered packet with three surveys (NDISS, the Measure of Job Satisfaction, and the demographics questionnaire), a $1 bill as a token of appreciation, and a pre-stamped return envelope. The in troduction letter in the survey packet included notice of a second repeat surv ey (NDISS) to follow. The follow-up postcard was sent out one week after the survey mailing and thanked participants who had returned the surveys and reminded those who had not yet returned the survey. The first survey deadline was two weeks from survey mailing.

PAGE 36

28 The second mailing series of Phase Tw o was sent 2.5 weeks after the first series to those participants who res ponded to the first survey. The second questionnaire packet consisted of anot her IRB-approved introduction letter serving as informed consent, a numbered NDISS survey, a $2 bill as a token of appreciation, and a pre-stamped return env elope. Participants were again asked to return the survey withi n two weeks. Another thank you/reminder postcard was sent out one week after the survey was mailed. The list of participants in Phase Tw o were numbered and stored on a disk in a locked location. Numbered surveys were used to collect test-retest data only, with no linking of personal id entification information. Data were entered into SPSS by the primary investigator. Phase Three Final survey packets were sent to 1,000 randomly-selected national APHON members. Mailings for Phase Three were similar to the multiple-contacts process in used in Phase Two, includi ng a pre-notification letter, a survey mailing, and a follow-up/thank you postcard. The pre-notification letter was sent out three days prior to the mailing of t he survey. The survey mailing contained an IRB-approved introduction letter serving as informed consent, a survey packet (containing the NDISS, t he Measure of Job Satisf action, and the demographics questionnaire), a pre-stamped return envelo pe to the primary investigator’s U.S. post office box, and a sheet of children’s stickers as a token of appreciation. There was no linking of personal identific ation information to the surveys. The

PAGE 37

29 deadline for survey return was three w eeks from the survey mailing. A thank you/reminder postcard was sent out 1.5 weeks after mailing the survey packets. Data was entered into SPSS by the primar y investigator. All original survey documents will be stored for three year s in a separate locked container. Data Analysis Phase One In the first part of Phase One, 20 APHON Listserv respondents were asked to rate the original list of 64 nursi ng interventions for those most frequently used in practice. Responses were summa rized per symptom for frequencies. In the second part of Phase One, a content validity index (CVI) was calculated by adding the responses (1, 0, or -1) for eac h question individually and then dividing by the number of raters (McMillan, 1990; McMillan et al., 1988). A total CVI was calculated by averaging the item-CVIs. A CVI of .80 represents adequate content validity (Polit & Beck, 2004). Phase Two The reliability of the NDISS was meas ured using a test-retest method. The test-retest evaluation was composed of two statistics (correlation and percentage-of-agreement) because of the differ ent types of questions used in the NDISS. Correlations were used to analyze the questions about nurses’ distress, the number of interventions used to treat the symptoms, and the nurses’ perceived efficacy of those nursing interven tions. A higher correlation coefficient

PAGE 38

30 demonstrates a more stable instrument greater than .70 is considered satisfactory (Polit & Beck, 2004). A percentage-of-agreement statistic (the average of each participant’s number of items with the same answers in both surveys) was used for the questions about the presence of patients’ symptoms and the type of nursing interventions used to treat each symptom. Items were scored as “1” if answers were the same between surveys per respondent. Items were scored as “0” if answers were different between surveys. Each of these questions was summarized by calculating the av erage of the agreement scores across participants. Phase Three Descriptive statistics were used to describe the study participants and summarize survey results. Pearson’s r (wit h a t-test of significance) was used to measure the correlations in test ing for the following hypotheses: 1. There is a positive relationshi p between the presence of distressing symptoms in pediatric oncology patients and the nurses’ distress from those symptoms. 2. There is an inverse relationship bet ween the nurses’ perceived effectiveness in treating patients’ symptoms and t he nurses’ distress from patients’ symptoms.

PAGE 39

31 3. There is an inverse relationship betw een the number of nursing interventions used to treat these symptoms and t he nurses’ distress from patients’ symptoms. 4. There is an inverse relationship between nurses’ distress and nurses’ job satisfaction. Demographic and nurses’ practice characte ristics were examin ed as covariates. Hierarchical regression analyses were used to test the following hypotheses: 5. The nurses’ perceived effectiveness of nursing interventions acts as a mediator between patients’ sym ptoms and nurses’ distress. 6. The quantity of nursing interventions ac ts as a mediator between patients’ symptoms and nurses’ distress. Hierarchical regression analysis was also used to examine the effectiveness of the study’s model in predicti ng overall job satisfaction.

PAGE 40

32 Chapter IV Results The purpose of this study was to examine the relationships between pediatric oncology nurses’ symptom m anagement of patients’ most distressing symptoms and job satisfaction. Chapter IV presents the results from each phase of the study. Phase One The focus of Phase One was to examin e the content validity of the newly developed Nurses Distress and Interventions for Symptoms Survey (NDISS). For the first part of Phase One, 20 nurses volunteered and were sent a list of 64 nursing interventions recommended in the lite rature for use in managing seven of pediatric oncology patient’s most di stressing symptoms. Twelve nurses responded (60%). Data was entered into E xcel by the primary investigator and descriptive statistics were used to summari ze the data. From t he original list of 64 interventions, the 35 most frequently selected interventions were chosen for inclusion in further NDISS versions. The overall mean number of nursing interventions chosen by the nurses per symptom was 19.63 interventions with a standard deviation of 5.6 – therefore 35 in terventions approximately represented the inclusion of three standard deviations of interventions.

PAGE 41

33 For the second part of Phase One, six of the seven CVI experts returned a completed NDISS evaluation form (86%). Data were entered in SPSS by the primary investigator. The CVI scores ranged from .75 to 1.0. The overall CVI of the NDISS (n=6) was .88 (SD .11). Phase Two The focus of Phase Two was to exam ine the reliability of the NDISS, as well as pilot test the multiple mailing technique and surveys to be used in Phase Three. One hundred pediatric oncology nurses were sent the first round of survey mailings. Sixty nurses (60%) responded to t he first survey. These 60 nurses were sent the second survey for test-retest, and 46 (77%) of t hem responded, yielding an overall response rate of 44% (n= 44). Two surveys were excluded because they did not meet inclusion criteria, namel y not working directly in patient care and too little experience in pediatric oncology (less than 6 months). The final sample of 44 nurses was predominately female (98%) and white (98%) with a mean age of almost 43 year s (Tables 1 and 2). The nurses had been practicing nursing a mean of about 20 years, with an average of 17 years in pediatric oncology. Geographically, nearly 40% of the nurses practiced in the South Atlantic division and another 25% pr acticed in the Mid-Atlantic division. The nurses predominately functioned as hospital inpatient (52%) staff nurses (55%) using their licensure as a Register ed Nurse (66%) to take care of children and adolescents (73%) (Table 3). Over one-th ird of the nurses (36%) worked in a

PAGE 42

34 Magnet-credentialed facility. Approximately 59% of nurses reported having a Bachelor’s or Master’s degree. Table 1. Summary of Phase Two Nurses’ Age and Years in Practice Demographic variable N Mean SD Age 43 42.7 10.2 Years in nursing 44 19.6 11.2 Years in pediatric oncology 44 17.0 10.3 Note. SD = standard deviation. The test-retest reliability score was calc ulated by two statistics: correlation and percentage of agreement. Correlations were used to assess the questions which addressed the nurses’ distress, the number of nursing interventions used, and the perceived effectiveness of nursing interventions. Nurses’ distress was the least reliable category between the fi rst and second surveys (r=.42; p=.01; n=43); number of interventions used per symptom had a correlation of .58 (p=.00; n=43), and perceived effectiveness of nursing interventions was the most stable between surveys (r=.72; p=.00; n=43). The average correlation between first and second NDISS surveys for these questions was .57.

PAGE 43

35 Table 2. Summary of Phase Two Demographics Demographic variable Frequency Percentage Gender Female 43 98 Race Caucasian/White Asian/Asian-American 43 1 98 2 Primary practice location South Atlantic Mid-Atlantic New England East North Central West North Central Other 17 11 4 4 4 4 39 25 4 4 4 4 Nurses highest level of education* Bachelor’s Associate’s Master’s Diploma Other 15 12 11 5 1 34 27 25 11 2 Note. N=44 *Totals may not equal 100% due to rounding error.

PAGE 44

36 Table 3. Summary of Phase Two Nursing Practice Characteristics Demographic variable Frequency Percentage Primary practice setting Hospital inpatient Hospital outpatient Home care Other 23 19 1 1 52 43 2 2 Primary position* Staff nurse Advanced practice (NP, CNS) Nurse manager/Administrator Educator 24 16 2 2 55 36 5 5 Nursing degree currently being used* Registered Nurse (RN) Nurse Practitioner (NP) Clinical Nurse Specialist (CNS) 29 14 1 66 32 2 Certified nurses 29 66 Work in Magnet facility No Yes Currently applying for Magnet status 22 16 6 50 36 13 Age of patient population Children and adolescents Children, adolescents and adults 32 12 73 27 Note. N=44. *Totals may not equal 100% due to rounding error.

PAGE 45

37 The remaining NDISS questions were examined for the percentage-ofagreement between surveys. The percentage-of-agreem ent between surveys for presence of symptom in patients was .92 (SD 0.17; n=44); and agreement for type of nursing intervention used to treat the symptom was .7 4 (SD 0.15; n=44). The overall average percentage of agreem ent for these questions was .83 (SD 0.12; n=44). Phase Three The focus of Phase Three was to examine symptom management and the sequelae of offering this management am ong a national sample of pediatric oncology nurses. Data are presented by a summary of general findings, followed by testing of the study’s hypothes es and model based on study findings. Sample Descriptors Survey packets consisting of t he NDISS, the MJS, and demographic information were sent to 1,000 pediatric oncology nurses nationally. Five hundred twenty-six (53%) pediatric oncology nurses returned the research surveys; 509 of these surveys were eligible for inclusio n in this study. Of the seventeen surveys not eligible for study inclusion, th irteen respondents stated they did not have direct care with patients at the time of survey comp letion, and four surveys did not complete the question about having direct patient care. Phase Three eligible respondents were largely female (98%), white (88%), and had a mean age of 40 years and had worked in pediatric oncology for a mean of 11.6 years (Tables 4 and 5). Tw enty-two percent of nurses were from

PAGE 46

38 the Pacific division, followed by South Atl antic (16%), Mid Atlantic and East North Central (13% each). Eighty-two percent of the nurses had a Bachelor’s or Master’s degree in nursing. Table 4. Summary of Phase Three Nurses’ Age and Years in Practice Demographic variable N Mean SD Age 500 40.0 10.5 Years in nursing 506 15.6 10.5 Years in pediatric oncology 509 11.6 8.3 The nurses were predominately hospit al inpatient (59%) staff nurses (70%) using a Registered Nurse (RN) licens e (78%) to provide care to children and adolescents (62%) (Table 6). The nurses were typically certified (76%) and just under a third worked in a M agnet-credentialed facility (32%). NDISS results of the presence of sym ptoms are summarized by frequency of symptom (Table 7). The average number of symptoms reported as present was 6.0 (SD 1.3). Pain was the most commonly reported symptom; trouble sleeping was the least common. NDISS resu lts for nurses’ distress, number of nursing interventions used, and perceived effectiveness of nursing interventions are presented as averages across sympt oms (Table 8). Nurses’ distress was greatest with trouble sleeping and lowest with hair loss. The overall average number of nursing interventions used to treat each symptom was 12.7; the greatest number was used to manage pai n; the least number was used to

PAGE 47

39 manage hair loss. Nurses reported pain as the most effectively treated symptom; fatigue was perceived as the least effectively managed. Table 5. Summary of Phase Three Demographics Demographic variable N Frequency Percentage Gender Female 508 497 98 Race Caucasian/White Asian/Asian-American Black/African-American Native American/Pacific Islander Other 508 446 24 13 6 19 88 5 3 1 4 Primary practice location* Pacific South Atlantic Mid-Atlantic East North Central West South Central New England West North Central East South Central Mountain 509 113 81 68 67 44 43 43 29 21 22 16 13 13 9 8 8 6 4 Nurses highest level of education* Bachelor’s Master’s Associate’s Diploma Other 509 268 147 66 23 5 53 29 13 5 1 *Totals may not equal 100% due to rounding error.

PAGE 48

40 Table 6. Summary of Phase Three Nursing Practice Characteristics Demographic variable N Frequency Percentage Primary practice setting Hospital inpatient Hospital outpatient Physician office/Private practice Other 509 299 180 18 12 59 35 4 2 Primary position Staff nurse Advanced practice Nurse manager/Administrator Educator Other 509 356 103 22 10 18 70 20 4 2 4 Nursing degree currently being used Registered Nurse (RN) Nurse Practitioner (NP) Clinical Nurse Specialist (CNS) Other 509 396 88 20 5 78 17 4 1 Certified nurses 505 382 76 Work in Magnet facility No Yes Currently applying for Magnet status 508 185 183 160 36 32 32 Age of patient population* Children and adolescents Children, adolescents and adults Other 509 313 192 4 62 38 1 Note. Sample size may vary according to respondents’ missing data. *Totals may not equal 100% due to rounding error.

PAGE 49

41 Table 7. NDISS Phase Three Summary Results of Symptoms Present Patient symptom present Frequency Percentage Pain 479 98 Nausea/Vomiting 474 97 Hair loss 455 93 Worry 435 89 Fatigue 421 86 Mouth sores 401 82 Trouble sleeping 328 67 Note. N=489. Table 8. NDISS Phase Three Summary Results Nurses’ distress (N=489) Number of nursing interventions used (N=435) Perceived effectiveness of interventions (N=414) Patient symptom Mean SD Mean SD Mean SD Trouble sleeping 3.4 2.0 11.6 7.3 2.4 0.8 Mouth sores 3.3 1.5 10.6 7.1 2.6 0.8 Worry 3.1 1.4 14.7 7.1 2.4 0.8 Pain 3.0 0.9 18.6 7.8 3.0 0.7 Nausea/Vomiting 2.9 1.0 13.0 7.4 2.9 0.7 Fatigue 2.7 1.6 11.3 7.4 2.1 0.8 Hair loss 1.8 1.6 9.1 5.2 2.4 1.1 Overall mean 2.9 0.8 12.7 6.1 2.5 0.5

PAGE 50

42 A complete ranking of interventions overall and per symptom can be found in Appendix B. The top five most frequently used interventions across all symptoms were (in order): emotional support, encourage family involvement, active listening, family support, and educ ation. The five most commonly used interventions per symptom we re (in order of symptom wi th the largest number of interventions used): pain – pain-reducin g medication, distraction, emotional support, active listening, and encourage fa mily involvement; worry – active listening, emotional support, encourage fam ily involvement, family support, and psychosocial support for patient; nausea/vomiting – nausea-reducing medications, anxiety-reducing medication s, distraction, emotional support, and encourage family involvement; trouble sle eping – adjust nighttime sleep regimen, sleep-inducing medications, reduced sl eep interruptions, anxiety-reducing medications, and relaxation; mouth sore s – mouth care/hygiene, pain-reducing medications, nutrition, encourage family involvement, and education; fatigue – encourage family involvement emotional support, assist with physical needs, reduced sleep interruptions, and adjust nighttime sleep regimen; and hair loss – emotional support, active listening, fam ily support, education, and anticipatory guidance. A summary of the Measure of Job Sati sfaction (MJS) results are available in Table 9. The overall score of nurses’ job satisfaction was 3.9 (SD 0.5; range 15). The highest scoring subscales were Personal Satisfaction and Satisfaction

PAGE 51

43 with Standards of Care. The lowest scoring subscale was Satisfaction with Workload. Table 9. Measure Of Job Satisfaction (MJS) Results for Phase Three MJS Subscale Mean SD Personal Satisfaction 4.2 0.5 Satisfaction with Standards of Care 4.2 0.6 Satisfaction with Prospects 4.0 0.6 Satisfaction with Professional Support 3.9 0.7 Satisfaction with Training 3.6 0.8 Satisfaction with Pay 3.6 0.9 Satisfaction with Workload 3.5 0.7 Overall Satisfaction Score 3.9 0.5 Note. N=508. Scores on the MJS ranged from 1 (v ery dissatisfied) to 5 (very satisfied). Bivariate Correlations Tabled correlations were used to evaluat e four of the six study hypotheses (Table 10). Two of the study’s hypotheses were rejected as relationships among the variables behaved conversely to the relationship expected. The first hypothesis proposed that as the presence of distressi ng symptoms in pediatric oncology patients increased, so did nurses’ distress from those symptoms. This hypothesis was rejected as there was in fact a significant inverse relationship found between these variables (r= -.67, p=.00). Rather, as the number of symptoms increased, nurses’ distress decreased.

PAGE 52

44 The second hypothesis proposed that as nurses’ perceived effectiveness in treating patients’ symptoms decreased, then nurses’ distress from patient symptoms would increase. This hypothesis was rejected as there was a positive correlation of .12 (p=.01) between nurses’ perceived effectiveness of interventions and nurses’ distress from symptoms. As the perceived effectiveness of nursing interventions increased, so did nurses’ distress. Two of the study hypotheses were not supported as there were no significant correlations among variables. The third hypothesis proposed an inverse relationship between the number of nursing interventions used to manage symptoms and the nurses’ distre ss from patients’ symptoms. This hypothesis was not support ed as there was no significant correlation between these variables (r = -.02). The fourth hypothesis proposed an inverse relationship between nurses’ distress and nurses’ job sa tisfaction. This hypothesis was not supported as the correlation was .04 and non-significant. Regression Analyses Regression analyses were used in addr essing the final two hypotheses. Correlations between variables were measured to analyze for potential covariates. Table 11 presents a correlation matrix of the dependent variables (job satisfaction and nurses’ distress), demographic covariates (race, geographic location) and nursing practice covariates (primary practice position and years in pediatric oncology) which were found to have significant relationships.

PAGE 53

45 Table 10. Correlations Among Main Study Variables Variable Job satisfaction Nurses’ distress Patients’ symptoms Perceived effectiveness Number of nursing interventions Job satisfaction -Nurses’ distress .04 -Patients’ symptoms .01 -.67***-Perceived effectiveness .16*** .12** .06 -Number of nursing interventions .02 -.00 .18*** .25*** -Note. N=508. **p<.01. ***p<.001. Demographic variables with signific ant correlations to the dependent variables were analyzed by subgroups for more meaningful interpretation. For example, race was found to be signific antly correlated to the dependent variables for respondents answering “White/Caucasian”. This subgroup was then coded and included in the correlation matrix (T able 11). The same scenario was true for location. Location proved significant only for the subgroup “Pacific” therefore this subgroup was retained as a potential covariate.

PAGE 54

46 Nurses’ age, years in nur sing, and years in pediatric oncology were all significant with main study variables, but highly correlated with each other (nurses’ age with years in nursing: r= .88; nurses’ age with years in pediatric oncology: r= .73; and years in nursing with years in pediatr ic oncology: r= .81; all significant at p=.00). The greatest effect s on the regression equations were noted with years in pediatric oncology, therefor e this variable was retained among the three. Table 11. Correlation Table of Dependent Variables and Covariates Variable Job satisfactio n Nurses’ distress Race: White Pacific Staff nurse Years in pediatric oncology Job satisfaction -Nurses’ distress .03 -Race: White .10* -.11* -Location: Pacific .10* .08 -.25*** -Primary position: Staff nurse .20*** .02 -.09* .14** -Years in pediatric oncology .01 .10* .09 -.01 .28*** -Note. N=498. *p<.05. **p<.01. ***p<.001.

PAGE 55

47 The two variables regarding nurses’ pr actice status (nursing degree being used: RN vs. NP; and primary practice position: staff vs. advanced practice) were both significant with the main study’s va riables but again strongly correlated (r= .77, p=.00). When analyzed by subcategor y, the subgroup of staff nurses within primary practice position offered the gr eatest contribution to the analysis and was therefore retained. Hierarchical regression analyses we re used to analyze the remaining hypotheses. Based on the correlations in Ta ble 11, covariates were entered into the regression analyses by blocks (dem ographics variables and nurse practice variables) in order to remo ve the effects of those va riables from the equation. Prior to the addition of nursing intervent ions as mediators, the number of symptoms had a multiple correlation (R) of .63 and was able to predict 39% of nurses’ distress. The variable ‘years in pediatric oncology’ acted as a suppressor variable in that it did not significantly improve the regression models itself, but helped to improved the model in expl aining nurses’ distress overall. The fifth hypothesis examined the use of perceived effectiveness of nursing interventions as a mediator between patients’ sym ptoms and nurses’ distress and was not rejected. Adding perceived effectiveness significantly contributed to the regressi on equation above the effects of the covariates and the total number of patients’ symptoms pres ent (Table 12). The use of perceived effectiveness as a mediator allowed t he prediction of 41% of nurses distress (R=.64). The variable ‘years in pediatric onc ology’ acted as a suppressor variable

PAGE 56

48 in that it did not significantly improve the regression models itself, but helped to improved the model in explaini ng nurses’ distress overall. The final hypothesis examined the num ber of nursing interventions as a mediator between patients’ symptoms and nurses’ distress. This hypothesis was not rejected as it also contributed signi ficantly in explaini ng nurses’ distress, above the effects of the covariates and pat ients’ number of symptoms (Table 13). Including the number of nursing interventi ons as a mediator increased the R to .64, explaining 40% of nurses’ distress. Table 12. Regression Findings Evaluating Perceived Effectiveness Variable R2 change F change p Step 1 Race: White Years in pediatric oncology .09* .05 .03 6.36 .00 Step 2 Total number of symptoms present -.62***.43 303.46 .00 Step 3 Perceived effectiveness of nursing interventions .15*** .02 18.25 .00 Note Dependent variable: Mean Amount of Nurses’ Distress. *p<.05. **p<.01. ***p<.001.

PAGE 57

49 Table 13. Regression Findings Evaluating Quantity of Nursing Interventions Variable R2 change F change p Step 1 Race: White Years in pediatric oncology .10** .04 .03 6.35 .00 Step 2 Total number of symptoms present -.64***.37 303.46 .00 Step 3 Number of nursing interventions (quantity) .11** .01 9.78 .00 Note Dependent variable: Mean Amount of Nurses’ Distress. *p<.05. **p<.01. ***p<.001. Finally, hierarchical regression was used to analyze the effectiveness of the study’s model in explaining overall job satisfaction. Analysis was performed using SPSS REGRESSION and SPSS FREQUENCIES for evaluation of assumptions. Table 14 presents the anal ysis findings, including the standardized regression coefficients ( ), and with the addition of each block of variables (demographics, nursing practice, and st udy variables) the change in R2 the change in F, and the significance of the change in F. Race (white/Caucasian) and location (Pacific division) were included in block 1 for demographic covariates; and block 2 contained primary practice position (staff nurse) and number of years in pediatric oncology as the nursing practice covariates. Each block contributed significantly in explai ning the study’s model. Years in pediatric

PAGE 58

50 oncology again behaved as a suppressor variable. Of the four main study variables, only “perceived effectivene ss of nursing interventions” offered a significant contribution to the final blo ck of the regression. However, without the other three variables, the ability of the model to predict job satisfaction was reduced. The fully mediated model was si gnificantly different than the null hypothesis with R= .33, adjusted R2=.09, and F (8,494) = 7.45, p=.00. Table 14. Regression Findings Evaluating Study Model on Job Satisfaction Variable R2 change F change p Step 1 Race: White Location: Pacific division .14** .16*** .03 6.73 .00 Step 2 Staff nurse Years in pediatric oncology -.23*** .07 .05 12.50 .00 Step 3 Total no. symptoms present Nurses’ distress Perceived effectiveness No. nursing interventions .04 .02 .19*** .07 .04 4.84 .00 Note. Dependent variable: Overall job sati sfaction. RN = registered nurse. ***Contributed significantly to the model when a ll three blocks included at the level of p<.001.

PAGE 59

51 Chapter V Discussion This chapter discusses the results presented in Chapter Four. Discussion is presented by phase, and concludes with a study summary including implications for practice. Phase Thr ee discussion includes the evaluation of survey findings, the hypotheses, and the study model. Phase One The purpose of Phase One was to test the content validity of the newly developed Nurses’ Distress and Interventions for Symptoms Survey (NDISS). In the first part of Phase One, the Associ ation for Pediatric Hematology/Oncology Nurses (APHON) nurses who responded from the Listserv gave useful feedback to assist in determining which nursing in terventions were most commonly used. This list was used to reduce the number of nursing interventions listed on the NDISS. However, some of the intervent ions removed during this stage were repeatedly written in on the Phase Two and Three surveys by nurses, including acupuncture, the use of cold and heat, massage, and use of wigs/hats. While acknowledging the burden of a long list of interventions, the longer list may have provided a richer description of intervent ions being used in subsequent samples (e.g. Phases Two and Three). With an obviously limited sample of 12 nurses, it is

PAGE 60

52 also possible that the nurses who responded from the Listserv were different than APHON nurses who did not respond or are not members of the Listserv at all. Ideally, validity would have been exam ined by comparing the patients’ statement of presence of symptoms and comparing them with the nurses’ perception of presence of symptoms. However, given time and financial constraints, content validity was seen as the most appropriate option. In part two of Phase One, the content validity i ndex (CVI) results for the NDISS (.88) demonstrated excellent cont ent validity for the newly developed instrument. Suggestions by experts during the CVI-portion of Phase One generally concerned the choice of patient symp toms included in the NDISS. As these symptoms were based on a literature re view of those most distressing to patients, the original seven symptoms were retained. The questions for each symptom were designed to be consistent between symptoms in order to make the survey easier for respondents. T here were, however, comments in both Phase Two and Phase Three that emphasiz ed the difficulty in assessing the ability to “treat” the symptom of hair loss. Future versions of the NDISS may consider modifying the questions to refl ect the management of patients’ distress from the symptoms, more than the symptom itself. Phase Two The purpose of Phase Two was to bot h examine the reliability of the NDISS and to pilot test the survey packe t and the multiple contact design. Test-

PAGE 61

53 retest reliability was chosen as the most suitable option to test NDISS reliability due to the nature of the questions and the given time and financial constraints. The NDISS questions each regarded different constructs (for example, nurses’ distress or perceived effectiveness of in terventions), therefore a measure of internal consistency was not appropriate. Cr iteria for the appropr iate use of testretest stability include the presence of t he same test forms, the same subjects, and the same situations. The first two crit eria were accomplished in Phase Two. Yet it became apparent based on nurses’ comments in Phases Two and Three that varying patient and work situations may have contributed to error variance in the reliability of the NDISS. Nurses wrote in the comments about particularly difficult patients or work assignments and the effect of those conditions on the nurses’ responses. Researchers interest ed in using the NDISS should consider modifying the questions by asking respondent s to identify a specific patient for consideration in answering the questions although that approach may limit the generalizability of findings and en courage polarity of responses. Phase Two participation was acceptable; however, the reliability statistics were not ideal. The percentage-of-agreem ent statistics were adequate; of the correlations, only the questions about perceived effectiveness of nursing interventions approached the acceptable minimum standards for reliability (r=.72). Nurses’ distress was the leas t stable question (r=.42). Some of the written-in comments indicated that nurses had a difficult time assessing their own distress in relation to patients’ symptoms. In addition, the choice of wording for

PAGE 62

54 these questions may have contributed to the error variance of this question type. For example, nurses may have had c onfusion in answering the question according to managing the symptom ve rsus managing the distress from the symptom. Ultimately, these low reliability findings may have compromised Phase Three findings as error variance is invers ely proportionate to re liability. With low reliability, there is far greater in fluence of error in the results. The demographics of Phase Two were skewed by geographic location as obvious when comparing the percentage of participants from the east and west coasts between Phases Two and Three. In distribution of the first mailing of Phase Two the proposed-randomized database of national pediatric oncology nurses had very few nurse addresses from the western U.S. This issue was addressed with the Association of Pediat ric Hematology/Oncology Nurses and remedied between Phases Two and Thr ee by the generation a new randomized list of 1,000 national members. Theref ore, Phase Three reflected a true randomized national sample. This comp romise in geographic representation should not have affected the reliability statistics. APHON allows membership to nurses with at least an RN license; therefore practical nurses were not expec ted to respond to the survey. As noted also in Phase Three, a large percentage of nurses had a Bachelor’s or Master’s degree. The lower number of Associate’ s degree and Diploma nurses in this study’s samples may indicate a di fference between nurses belonging to a professional organization and non-mem ber nurses. The amount that this

PAGE 63

55 difference may influence survey findings is unknown. Samples in future studies should attempt to include non-members as well. Regarding the multiple contacts design recommended by Dillman (2007), there was a very positive response fr om participants – in both Phase Two and Three. The nurses wrote comments that they appreciated the pre-survey letter informing them of the arrival of the surv ey in the mail in the next few days. The follow-up contact also appeared to be effect ive as there was a surge of surveys following each follow-up postcard mailing. Phase Three Sample Descriptors The purpose of Phase Three was to examine the relationships between pediatric oncology nurses’ symptom management and job satisfaction. This section discusses the demographics of Phase Three respondents, as well as results from the NDISS and Meas ure of Job Satisfaction. A response rate of over 50% greatly exceeded expectations, and is likely related to the use of the multiple c ontacts design. Similar to Phase Two, respondents were mostly Caucasian female s. The percentage of Caucasians did decrease from 98% in Phase two to 88% in Phase Three, likely to due the increased representation from the Western U.S. Phase Three nurses were on average a few years younger (Phase Two mean 42.7 years, Phase Three 40 years) and had been working in pediatric on cology an average of over 5 years less (Phase Two mean 17 years, Phase Th ree mean 11.6). These findings also

PAGE 64

56 may be related to the exp anded geographic inclusion wit h Phase Three. Again, as in Phase Two, most nurses held a Ba chelor’s or Master ’s degree (82%). The rate of certified nurses in Phase Three ( 76%) is likely not representative of typical pediatric oncology nurses who may not be members of APHON, and this may have affected the generalization of survey findings. The NDISS was based on pediatric onc ology patients’ most distressing symptoms from the lit erature, which were (in or der) fatigue, pain, decreased appetite, nausea/vomiting, hair loss, isolat ion, worry, fear, m outh sores, trouble with mobility, trouble with relationships, and trouble sleeping (Collins et al., 2002; Drake et al., 2003; Enskar et al., 1996; H edstrm et al., 2003; Hicks et al., 2003; Hinds et al., 1992; Jalmsell et al., 2006; McCaffrey, 2006; Moody et al., 2006; Novakovic et al., 1996; Wolfe et al., 2000). Decreased appetite, isolation, fear, and trouble with mobility and relationships were excluded from the NDISS as it was felt these symptoms would be less tangi ble and therefore more difficult to assess. In retrospect, hair loss and trouble sleeping were at least difficult and may have been replaced by the more dist ressing symptoms such as isolation and fear. Frequency of symptoms is notably differ ent than symptom distress. A child may have the presence of a symptom but not feel bothered by that symptom. According to the literature, some of the most frequently o ccurring symptoms in pediatric oncology patients are fatigue, nausea, difficulty eating, fever, mucositis, pain, and hair loss (Drake et al., 2003; Williams, Schmideskamp, Ridder, &

PAGE 65

57 Williams, 2006). This study focus ed on the symptoms considered most bothersome or distressing to childr en with cancer. Surprisingly, fatigue and mouth sores were among the most frequent and most distressing symptoms, yet were rated by nurses as occurring less often than the other symptoms (86% and 82%, respectively). Nurses reported an average of six of the seven symptoms as present in their patients within the past month. Pain was reported as pres ent by 98% of the nurses, which is consistent with the literat ure in terms of frequency. This finding may also have to do with the design of t he survey listing pain first among the symptoms. Consistent with the literature about patients’ distress, nurses were also most distressed by pain; however, nur ses in this study reported the greatest perceived effectiveness of nursing intervent ions with pain. This is somewhat contrary to the literature review where nurses, particularly hospice nurses, felt that pain was nearly impossible to cont rol (Papadatou et al., 2002; Papadatou et al., 2001). Future studies might examin e patient and nurse perception of pain concurrently to determine the accuracy of nurses’ assessment of patient’s perception of pain. Nurses reported the highest levels of distress with patient worry. Nursing care of worry is time-consuming and somewhat elusive. The most common nursing interventions for treating worry were active listening, emotional support, and encouraging family involvement. Given today’s fast-paced hospital routine

PAGE 66

58 with a large number of nursing responsibilities, it is no wonder that nurses feel difficulty in caring for patients with worry. Surprisingly, nurses generally reli ed heavily on medication-based therapy, despite the emphasis on nursing interventions. Based on write-in comments on the surveys, it appears that nurses feel that medical management is quite similar to nursing management. For example, the most commonly occurring written-in intervention for fatigue was transfusion with packed red blood cells – clearly a medical intervention; mucositis frequently had electrolyte supplementation and specific medication-based mouthwashes wr itten-in as interventions. There was certainly a large number of advance pr actice nurses who would be capable of medically managing patients, yet the co mments were not limited to only those advance practice nurses. These findings are useful in helping to guide future study – emphasis might be placed on distress and perceived effectiveness of interventions when comparing nursing interventions vers us medical management by nurses. Perceived effectiveness might also prove more useful if studied per intervention rather than collectively across symptom s. Many nurses’ comments addressed the difficulty in “making” the physician write appropriate dosages for medications in order to better manage symptoms. Fu ture NDISS revisions might include an intervention that addresses the nurse’s advocacy for patients with physicians. Nurses may feel more comfortable usi ng more nursing-based interventions than in trying to control medi cal management indirectly. Emphasis might be placed on

PAGE 67

59 encouraging the development and effectivene ss of nursing interventions despite a highly “medical-ized” environment. Fatigue, although rated highly in frequen cy and distress by patients in the literature, was rated as pr esent by only 86% of the nur ses, and caused relatively little distress in the nurses (mean distress from fatigue: 2.7; range: 1 “not at all” to 5 “very much”). Fatigue was reported as having the least effectiveness in being treated by nursing interventions. Nurs es primarily used nursing-based interventions for treatmen t, although the general recomm endations for fatigue in the literature were not t he most frequently used interv entions. For example, the National Comprehensive Care Network publis hes guidelines for supportive care. The recommended management of fatigue includes education, energy conservation and activity clustering, dist raction, exercise, re laxation, nutritional adjustments, sleep hygiene and family involvement (National Comprehensive Cancer Network, 2007b). Of these reco mmendations, only the last two were present in the top five interventions for treating fatigue in this survey. Trouble sleeping was reported as t he least common symptom (67%). Interestingly, trouble sleeping was rated as the most distressing symptom to the nurses who felt it was present in their pat ients. The effect of trouble sleeping may be specific to nurses working during the eveni ng or night shift. Future versions of the NDISS might include a question to deter mine which shift the nurse primarily works. Trouble sleeping fell mid-range fo r the number of interventions used to treat this symptom, and nurse s were generally indifferent as to the effectiveness

PAGE 68

60 of these interventions (mean 2.4; range from 1 “ not at all” effective to 5 “very much” effective). Two of the top five nursing interventions involved for trouble sleeping included the use of medications (f or sleep and anxiety). While there are scant recommendations for insomnia or diff iculty sleeping in children with cancer, adult management recommends highly nur sing-related activities, including promotion of exercise, nutritional modifica tions, establishing routines, relaxation, and positioning (Valdres, Esca lante, & Manzullo, 2001). Hair loss was generally present in patients (93%), but nurses were relatively unbothered by its presence (mean distress from hair loss: 1.8; range: 1 “not at all” to 5 “very mu ch”). Nurses responded that th ey felt that interventions were effective in treating the symptom of hair loss. Hair loss is an obvious example of the confusion question wording about managing the symptom – nurses may have been unclear about ans wering the question according to treating the hair loss itself or the child’s distress from the hair loss. The most commonly written-in interventions were fo r wigs, hats, and referrals to programs assisting with these devices. These interv entions were not included in the NDISS because while these interventions ranked highly for management of hair loss (16th), overall, these interventions were ranked low due to the lack of application across other symptoms. The Measure of Job Satisfaction (MJS) results yielded a relatively high overall job satisfaction (mean 3.9; range 1 “ver y dissatisfied” to 5 “very satisfied”). These findings are consistent with previous ly published studies of job satisfaction

PAGE 69

61 in pediatric oncology nurses using the MJ S (Hinds et al., 1998; Hinds et al., 2003). There is potential bias in these re sults if the nurses who completed and returned the survey were more moti vated and happier with work than their counterparts. As there was no way to track the profiles of the nurses who did not return the survey, this potentia l cannot be investigated further. Overall job satisfaction was signific antly negatively correlated with both primary practice position as a staff nurse and among nurses practicing with an RN; and positively correlated with a prim ary practice position as an advanced practice nurse (CNS/NP) and among nurses using a Nurse Practitioner’s degree. That is, staff nurses or RN’s had lowe r job satisfaction scores than advanced practice nurses or NP’s. Upon closer exam ination, staff nurses were significantly negatively correlated in every job satisfaction subscale. This finding may be related to the issue of control regarding patient care and the ability to directly order medical interventions. According to comments, nurses were at times frustrated with not being able to c hange the medical management of some symptoms. Additionally, writ e-in comments from staff nurses often addressed the over-worked and under-paid conditions in the hospital/outpatient environment. These findings may be not be specific to pediatric oncology nursing, but rather consistent across nursing specialties given the current healthcare management structure and focus on cost-reduction. Years in pediatric oncology nursing wa s significantly positively correlated with the “Satisfaction with Pay” subsca le. Ideally nurses are collecting pay

PAGE 70

62 commensurate with their experience as this finding might illustrate. This may also be related to the increased expectati ons among the younger generations of nurses (the Generation X and Millenials) in having pay and work conditions competitive with other tech-savvy industry positions (Sherman, 2006). Nurses in Phase Three of this st udy demonstrated t he highest scores in the subscales of personal satisfaction (m ean 4.2; range 1-5) and in satisfaction with standards of care (mean 4.2). Numerous comments on the surveys attest to the fact that nurses care exceedingl y about, and take great pride in, offering excellent patient care. This is consistent with the literature t hat nurses find great meaning in the relationships with pati ents and families (Bertero, 1999; ClarkeSteffen, 1998; Cohen et al., 1994; Cohen & Sarter, 1992; Fall-Dickson & Rose, 1999; Grunfeld et al., 2005; Haberman, Germ ino, Maliski, Stafford-Fox, & Rice, 1994; Olson et al., 1998; Papadatou et al ., 2002). But this care comes at an expense – the lowest rating subscale was satisfaction with workload (mean 3.5). Again, frequent comments were writt en in the survey about the stress of “squeezing everything in” and feeling overworked. Upon closer examination, no single j ob satisfaction subscale was more predictive of any of the main study va riables than overall job satisfaction. Therefore, the decision was made to retain overall job satisfaction as representative of this concept when testing hypotheses.

PAGE 71

63 Bivariate Correlations The main study variables all proved to have significant correlations, though not in the anticipated directions. T he first hypothesis interestingly found an inverse relationship between presenc e of distressing sym ptoms and nurses’ distress as compared to the positive relationship proposed. The greater the number of symptoms present, th e less the distress felt by the nurse. This finding is unexpected and quite substantial in effect (r=-.67, p=.00). Upon closer inspection of this phenomenon, the trend wa s linear, with no special effects noted according to number of symptoms. Simply the fewer the symptoms, the greater the distress; and the greater the number of symptoms, the less the distress. This effect be related to the over whelming and emotionally-challenging task of managing multiple patient symptom s. If a patient presents with a greater number of symptoms, then the severity of those symptoms may also be greater. Therefore, in order to function effectiv ely as a nurse and to emotionally protect oneself, nurses may need to dissociate some what from the patient’s symptoms. As nurses become more task-oriented in managing patients’ symptoms, perhaps their distress decreases. This is support ed by a recent qualitative study finding that for managing symptoms in oncology patients, often nurses chose those symptoms that were easier to trea t and easier to measure improvement (Blomberg & Sahlberg-Blom, 2007). A dditionally, the qualitative study emphasized the difficulty in treating the less physical or tangible symptoms, for example worry or anxiety.

PAGE 72

64 Another proposition may be related to the idea that nurses take comfort and feel useful in performing tasks to help manage symptoms. As the number of symptoms increases, there is more task-wo rk to attempt to manage the multiple symptoms. The sense of staying busy performing tasks to benefit patients may allow the nurse to feel that she/he is helping the patient and therefore feel less distressed. If the nurse knows of only a handful of nursing interventions to treat each symptom, when there are not many symptoms, the nurse depletes her/his perceived options in offering nursing care and therefore feels more distressed. The second hypothesis was also surprising and contrary to the hypothesized relationship as it reveal ed that the perceived effectiveness of nursing interventions for treating symptom s was associated with greater feelings of nurses’ distress instead of a decrease in distress. Nurses with greater distress are more likely to feel that nursing in terventions help manage the symptoms. The correlation is fairly weak (r=.12), however significant (p=.01). This also may be related to empathy. Nurses may feel that they are able to help manage patients’ symptoms with interventions, but still feel bothered by the inability to completely resolve the symptoms and t herefore feel more distressed. The previous hypothesis demonstrated the inverse re lationship between nurses’ distress and number of symptoms. There was however no significant relationship between the number of symptoms and per ceived effectiveness. In light of the fairly weak co rrelation between nurses’ distress and perceived effectiveness, it is possible t hat the results may be skewed due to the

PAGE 73

65 error variance in the NDISS instrument itself, or by virtue of a non-normal distribution of the variables. The mean amount of nurses’ distress was 2.9 (SD 0.8; range 0-4). Most nurses were unable to rate their di stress as “not at all”. This may be related to social desirability, or it may be related to the inherent sense of empathy common to pediatric onco logy nurses. The mean perceived effectiveness was 2.5 (SD 0.5; range 0-4), another distribution skewed to the left. As error variance increases, the result s may become compromised, and this may be such an example. Two additional hypotheses regarding nurse s’ distress were found to have no significant relationship, and therefor e the hypotheses were not supported. There was no significant relationship between the number of nursing interventions used and nurses’ distre ss, nor between nurses’ distress and nurses’ job satisfaction. Nurses’ distre ss does not appear to fit as hypothesized within the study model. Besides references to pediatric oncology nurses’ distress in caring for dying patients, there is little published research on which symptoms nurses find most distressing. A qualitativ e approach, such as grounded theory or phenomenology, may be useful in explori ng these concepts from the nurses’ perspective before revising the model and NDISS instrument. Future studies might examine nurses’ perceptions of the most distressing symptoms and compare these to patients’ most dist ressing symptoms. Perhaps the patients’ most distressing symptoms chosen for incl usion in the NDISS were not the most

PAGE 74

66 appropriate and therefore NDISS revisi ons might replace symptoms on the currently on the NDISS with ot her distressing symptoms. Perceived effectiveness of nursing interventions however appeared more helpful to this model than previously hypothesized with two significant relationships. As the number of sympto ms increased, so did the number of nursing interventions used (r=.18, p= .00); and as the number of nursing interventions increased, nurses’ per ceived effectiveness increased (r=.25, p=.00). Therefore, perhaps there is a cumulative effect of nursing interventions. Any one particular intervention may not work well, but a synergistic effect may be perceived when multiple nursing in terventions are used together. Regression Analyses The remaining two hypotheses that examined nursing in terventions as mediators in the model were retained. Both perceived effectiveness of nursing interventions (quality) and the number of nursing interventions (quantity) contributed to the prediction of nurses’ distress. Prior to the nursing interventions as mediators, the number of symptoms alone (after t he effects of demographic and practice covariates) was able to pr edict 39% of nurses’ distress. Each nursing intervention variable (number of interventions and perceived effectiveness) was able to significantly im prove predictability of nurses’ distress. By adding both nursing intervention variabl es, the model was able to predict 42% of nurses’ distress. Nursing intervent ions, both in quantity and quality, appear useful in predicting the work environm ent of pediatric oncology nurses.

PAGE 75

67 The hierarchical regression analysi s of the study model, including covariates, was only able to predict a sm all portion of job satisfaction (adjusted R2 = .09). This is not surprising given t he generally weak prediction offered in the original Stress Response Sequence Model (Hinds et al., 1998; Hinds et al., 2003). In addition, there are other issues involved in job satisfaction that are not addressed in the model, for example pay. The work done in the original SRSM showed promising results in the developm ent of role-related meaning to help explain reactions of stress in pediatric oncology nurses. Future studies might consider combining the st ronger variables from each study, including symptom management and perceived effectiveness of nursing interventions along with role-related meaning. The correlations among the main study variables modify the study model with patients’ symptoms, number of nursing intervent ions, and nurses’ distress correlating with nurses’ perceived effectiv eness of nursing interventions – instead of nurses’ distress. Nurses’ distress was also found to be unrelated to the number of symptoms present in patients. This again may be related to the potential error in measuring distress. Distress may not be a stable variable to assist in measuring satisfaction, as supported by the nurses’ comments about difficult patients affecting their res ponses. Perceived effectiveness of nursing interventions was the only variable with a significant correlation to nurses’ job satisfaction, and therefore it is not surp rising that this was the only main study variable to contribute signifi cantly to the study model.

PAGE 76

68 The covariates were interesting. W hen subcategorized, “White/Caucasian” race was significantly negatively corre lated with perceived effectiveness of nursing interventions, significantly negativel y correlated with nurses’ distress, and significantly positively correlated with job satisfaction (see Table 11). Among other demographic variables, “White/Caucasi an” was also significantly correlated negatively with living in the Pacific region and positively with y ears of experience in pediatric oncology. These findings may be skewed by the extremely large percentage (88%) of responde nts in the White categor y. Drawing conclusions from these findings seems di fficult as the relationships are relatively weak, though significant. This may be a simp le product of measurement error. Conclusions Although the study’s model requires m odification, the findings generated from this study will provide baseli ne data for researchers on the nurses’ perceptions of the presenc e of patients’ symptoms, the nurses’ distress from these symptoms, the nurs ing interventions most frequently used in symptom management of the most distressing sym ptoms, the perceiv ed effectiveness of nurses’ interventions, a summary of job satisfaction and demographics of a national sample of pediatric oncology nurses. The research study also provided baseline data about the way pediatric oncology nurses manage patients’ symptoms. Nursing interventions, both in quantity and quality, have led to interesti ng information. The number of nursing interventions and the perceived effect iveness of nursing interventions both

PAGE 77

69 proved significantly related to study vari ables. Future research might examine the impact of nursing interventions on nurses’ stress levels and the subsequent relationship to job satisfac tion. Also, studies might ex amine nurses’ choice of interventions and the relationship to the nurses’ perceived theoretical practice framework. Due to the lack of published literat ure on nurses’ distress and nursing interventions in the literature, it is difficult to compar e adult and pediatric populations. Nursing empathy and sense of satisfaction in patient, family, and coworker relationships seem consistent bet ween populations. However, it is unclear if pediatric and adult oncology nurses re spond differently to patients’ symptoms perhaps as a result of having parents involved, or due to the perceived vulnerability of children in general. Based on the findings from this st udy, future research may address opportunities in promoting particular ev idence-based nursing interventions for symptom management based on re search and/or guidelines The data from this research demonstrates which nursing interventions are being used. It is recommended that nurses review the resear ch and other forms of evidence from reputable sites in establishing sympto m management practice guidelines. For example, the National Comprehensive Ca re Network offers supportive care practice guidelines, many of which have s pecific pediatric interventions, online or in paper version, available for free (available at www.nccn.org ). Once a consistent symptom management plan is pr acticed, nurses may then be able to

PAGE 78

70 measure how well the nursing interventi ons are actually working and find opportunities for improv ement symptom management. Also evident from this study was t he lack of distinction between nursing and medical interventions in symptom management. Nurses may consider defining their role in pediatric oncology according to the strengths that they exclusively bring to symptom m anagement from a nursing perspective. Promoting the image and use of nursing interventions may provide more holistic care to patients as they are concurrently being managed by a medical team. Useful baseline data from this study has now paved the way for future studies to examine specific symptoms and/or nursing interventions for symptom management in pediatric oncology. In additi on this study helps to describe the distress of pediatric oncology nurses and the methods that these nurses use to manage their patients’ symptoms. Future theoretical work in pediatric oncology should include nursing interventions as an influence in the work environment of nurses.

PAGE 79

71 References American Cancer Society. (2006). Cancer Facts and Figures 2006. Retrieved February 5, 2007, from http://www.cancer.org/downloads/STT/CAFF2006PWSecured.pdf Anderson, K., Anderson, L. E., & Glanze, W. D. (1994). Mosby's medical, nursing, and allied health dictionary : illustrated in full color throughout St. Louis: Mosby, Edition: 4th ed. / re vision editor, Kenneth N. Anderson ; consulting editor and writer, Lois E. Anderson ; consulting and pronunciation editor, Wa lter D. Glanze. Antonarakis, E. S., Evans, J. L., Heard, G. F., N oonan, L. M., Pizer, B. L., & Hain, R. D. (2004). Prophylaxis of acut e chemotherapy-induced nausea and vomiting in children with cancer: What is the evidence? Pediatr Blood Cancer, 43 (6), 651-658. Association of Pediatric Hema tology/Oncology Nurses. (2007). Pediatric oncology nursing: Scope and standards of practice Glenview, IL: APHON. Baggott, C. R., Kelly, K. P., Fochtman, D., & Foley, G. V. (Eds.). (2002). Nursing Care of Children and Adolescents with Cancer (3rd ed.). Philadelphia: Saunders. Barnard, D., Street, A., & Love, A. W. (2006). Relationships between stressors, work supports, and burnout among cancer nurses. Cancer Nursing, 29 (4), 338-345. Bertero, C. (1999). Caring for and about cancer patients: identifying the meaning of the phenomenon "caring" through narratives. Cancer Nursing, 22 (6), 414-420.

PAGE 80

72 Blegen, M. A. (1993). Nurs es' job satisfaction: a meta-analysis of related variables. Nursing Research, 42 (1), 36-41. Blomberg, K., & Sahlberg-Blom, E. (2007). Closeness and di stance: a way of handling difficult situations in daily care. Journal of Clinical Nursing, 16 (2), 244-254. Bond, D. C. (1994). The meas ured intensity of work-rel ated stressors in pediatric oncology nursing. Journal of Pediatric Oncology Nursing, 11 (2), 44-52; discussion 53-44. Bulecheck, G. M., & Mc Closkey, J. C. (1999). Nursing Interventions: Effective Nursing Treatments (3rd ed.). Philadelphia: W.B. Saunders Company. Bulecheck, G. M., McCloske y, J. C., Titler, M. G. & Denehey, J. A. (1994). Report on the NIC Project: Nursing interventions used in practice. American Journal of Nursing, 94 (10), 59-64, 66. Clarke-Steffen, L. (1998). The meaning of peak and nadir experiences of pediatric oncology nurses: secondary analysis. Journal of Pediatric Oncology Nursing, 15 (1), 25-33. Cohen, M. Z., Haberman, M. R., Steeves, R., & Deatri ck, J. A. (1994). Rewards and difficulties of oncology nursing. Oncology Nursing Forum, 21 (8 Suppl), 9-17. Cohen, M. Z., Harle, M., Woll, A. M., Despa, S., & Muns ell, M. F. (2004). Delphi survey of nursing research priorities. Oncology Nursing Forum, 31 (5), 1011-1018. Cohen, M. Z., & Sarter, B. (1992). Love and work: oncology nurses' view of the meaning of their work. Oncology Nursing Forum, 19 (10), 1481-1486. Collins, J., Byrnes, M. E., Dunkel, I. J., Lapin, J., Nadel T., Thaler, H. T., et al. (2000). The measurement of sympto ms in children with cancer. Journal of Pain and Symptom Management, 19 (5), 363-377.

PAGE 81

73 Collins, J., Devine, T., Di ck, G., Johnson, E., Kilham, H., Pinkerton, C., et al. (2002). The measurement of symptoms in young children with cancer: the validation of the Memori al Symptom Assessment Scale in children aged 712. Journal of Pain and Symptom Management, 23 (1), 10-16. CureSearch. (n.d.). About CureSearch. Retrieved February 5, 2007, from http://www.curesearch.org/uploadedFiles/AboutUs/CureSearchfactshe et04.17.pdf de Carvalho, E. C., Muller, M., de Carval ho, P. B., & de Souza Melo, A. (2005). Stress in the professional practice of oncology nurses. Cancer Nursing, 28(3), 187-192. Dillman, D. A. (2007). Mail and internet surveys : the tailored design method : with new internet, visual, and mixed-mode guide (2nd ed.). Hoboken, N.J.: Wiley. Dossey, B. M., Keegan, L., & Guzz etta, C. E. (Eds.). (2005). Holistic nursing : a handbook for practice (4th ed.). Sudbury, MA: Jones and Bartlett. Drake, R., Frost, J., & Col lins, J. J. (2003). The symptoms of dying children. Journal of Pain and Symptom Management, 26 (1), 594-603. Elliott, G. R., & Eisdorfer, C. (1982). Stress and human health: Analysis and implications of research. A study by the Institute of Medicine. National Academy of Sciences. New York: Springer Pub. Co. Emery, J. E. (1993). Pe rceived sources of stre ss among pediatric oncology nurses. Journal of Pediatric Oncology Nursing, 10 (3), 87-92. Enskar, K., Carlsson, M., Hamrin, E., & Kreuger, A. (1996). Swedish Health care personnel's perceptions of dis ease and treatment-related problems experienced by children with cancer and their families. Journal of Pediatric Oncology Nursing, 13 (2), 61-70; discussion 71. Ergun, F. S., Oran, N. T., & Bender, C. M. (2005). Qua lity of life of oncology nurses. Cancer Nursing, 28 (3), 193-199.

PAGE 82

74 Fall-Dickson, J. M., & Rose, L. (1999) Caring for patients who experience chemotherapy-induced side effects: the meaning for oncology nurses. Oncology Nursing Forum, 26 (5), 901-907. Faul, F., & Erdfelder, E. (1992). GPOWER : A priori, post-hoc, and compromise power analyses for MS-DOS [Comput er program]. Bonn, FRG: Bonn University, Department of Psychology. Fitch, M. I., Bakker, D., & Conlon, M. (1999) Important issues in clinical practice: perspectives of oncology nurses. Canadian Oncology Nursing Journal, 9 (4), 151-164. Florio, G. A., Donnelly, J. P., & Zevon, M. A. (1998). The structure of workrelated stress and coping among oncolog y nurses in high-stress medical settings: a transactional analysis. Journal of Occupational Health Psychology, 3 (3), 227-242. Foley, G. V., & Ferguson, J. H. (2002). History, Issues, and Trends. In C. R. Baggott, K. P. Kelly, D. Fochtman & G. V. Foley (Eds.), Nursing Care of Children and Adolescents with Cancer (3rd ed., pp. 2-23). Philadelphia: Saunders. Grunfeld, E., Zitzelsberger, L., Coristi ne, M., Whelan, T. J., Aspelund, F., & Evans, W. K. (2005). Job stress and j ob satisfaction of cancer care workers. Psycho-Oncology, 14 (1), 61-69. Haberman, M. R., Germino, B. B., Maliski, S., Stafford-Fox, V., & Rice, K. (1994). What makes oncology nursing specia l? Walking the road together. Oncology Nursing Forum, 21 (8 Suppl), 41-47. Hedstrm, M., Haglund, K., Skolin, I., & von Essen, L. (2003). Distressing events for children and adolescents with cancer: child, parent, and nurse perceptions. Journal of Pediatric Oncology Nursing, 20 (3), 120-132. Helmrich, S., Yates, P., Nash, R., Hobman, A., Poulton, V., & Berggren, L. (2001). Factors influencing nurses' dec isions to use non-pharmacological therapies to manage patients' pain. Australian Journal of Advanced Nursing, 19 (1), 27-35.

PAGE 83

75 Hessig, R. E., Arcand, L. L., & Frost, M. H. (2004). The effects of an educational intervention on oncology nurses' attit ude, perceived knowledge, and selfreported application of co mplementary therapies. Oncology Nursing Forum, 31 (1), 71-78. Hicks, J., Bartholomew, J., Ward-Smith, P., & Hutto, C. (2003). Quality of life among childhood leukem ia patients. Journal of Pediatric Oncology Nursing, 20 (4), 192-200. Hinds, P. S. (2000). Testing the stress-re sponse sequence in pediatric oncology nursing. Journal of Pediatric Oncology Nursing, 17 (2), 59-68. Hinds, P. S., Fairclough, D. C., Dobos, C. L., Greer, R. H ., Herring, P. L., Mayhall, J., et al. (1990). Developm ent and testing of the Stressor Scale for Pediatric Oncology Nurses. Cancer Nursing, 13 (6), 354-360. Hinds, P. S., Quargnenti, A., Bush, A. J., Pratt, C., Fairclough, D., Rissmiller, G., et al. (2000). An evaluation of the impac t of a self-care coping intervention on psychological and clinical outco mes in adolescents with newly diagnosed cancer. Eur J Oncol Nurs, 4 (1), 6-17; discussion 18-19. Hinds, P. S., Quargnenti, A. G., Hickey, S. S., & M angum, G. H. (1994). A comparison of the stress--respons e sequence in new and experienced pediatric oncology nurses. Cancer Nursing, 17 (1), 61-71. Hinds, P. S., Quargnenti, A. G., & Wentz, T. J. (1992). Measuring symptom distress in adolescents with cancer. Journal of Pediatric Oncology Nursing, 9 (2), 84-86. Hinds, P. S., Sanders, C. B., Srivastava D. K., Hickey, S., Jayawardene, D., Milligan, M., et al. (1998). Testing t he stress-response sequence model in paediatric oncology nursing. Journal of Advanced Nursing, 28 (5), 11461157. Hinds, P. S., Srivastava, D. K., Randall, E. A., Green, A., Stanfor d, D., Pinlac, R., et al. (2003). Testing the revised stress-response sequence model in pediatric oncology nurses. Journal of Pediatric Oncology Nursing, 20 (5), 213-232.

PAGE 84

76 Jalmsell, L., Kreicbergs, U., Onelov, E. Steineck, G., & Henter, J. I. (2006). Symptoms affecting children with ma lignancies during the last month of life: a nation wide follow-up. Pediatrics, 117 (4), 1314-1320. King, M. O., Pettigrew, A. C., & Reed, F. C. (1999). Complement ary, alternative, integrative: have nurses kept pace with their clients? Medsurg Nursing, 8 (4), 249-256. Kushnir, T., Rabin, S., & Azulai, S. (1997). A descriptive study of stress management in a group of pediatric oncology nurses. Cancer Nursing, 20 (6), 414-421. Ladas, E. J., Post-White, J., Hawks, R. & Taromina, K. (2006). Evidence for Symptom Management in the Child With Cancer. Journal of Pediatric Hematology/Oncology, 28 (9), 601-615. McCaffrey, C. N. (2006). Major stressors and their effects on the well-being of children with cancer. Journal of Pediatric Nursing, 21 (1), 59-66. McMillan, S. C. (1990). Nurses' comp liance with American Cancer Society guidelines for cancer pr evention and detection. Oncology Nursing Forum, 17 (5), 721-726. McMillan, S. C., & Small, B. J. (2002) Symptom distress and quality of life in patients with cancer newly admitted to hospice home care. Oncology Nursing Forum, 29 (10), 1421-1428. McMillan, S. C., Small, B. J., Weitzner, M., Schonwetter, R., Ti ttle, M., Moody, L., et al. (2006). Impact of coping skills inte rventions with family caregivers of Hospice patients with cancer. Cancer, 106 (1), 214-222. McMillan, S. C., Williams, F. A., Chatfield, R., & Camp L. D. (1988). A validity and reliability study of two tools for assessing and managing cancer pain. Oncology Nursing Forum, 15 (6), 735-741.

PAGE 85

77 Moody, K., Meyer, M., Mancuso, C. A ., Charlson, M., & R obbins, L. (2006). Exploring concerns of children with cancer. Supportive Care in Cancer, 14(9), 960-966. National Comprehensive Cancer Network. (2007a). NCCN clinical practice guidelines in oncology: Antiemesis, V.1.2007 Retrieved April 12, 2007, from http://www.nccn.org/professionals /physiciangls/PDF/antiemesis.pdf National Comprehensive Cancer Network. (2007b). NCCN clinical practice guidelines in oncology: Cancer-related fatigue, V.2.2007 Retrieved April 21, 2007, from http://www.nccn.org/professiona ls/physiciangls/PDF/fatigue.pdf National Comprehensive Cancer Network. (2007c). NCCN clinical practice guidelines in oncology: Distress management, V.1.2007. Retrieved April 21, 2007, from http://www.nccn.org/professionals/ physiciangls/PDF/distress.pdf National Comprehensive Cancer Network. (2007d). NCCN clinical practice guidelines in oncology: Pedi atric cancer pain V.1.2006 Retrieved April 21, 2007, from http://www.nccn.org/professionals/physiciangls/PDF/pediatricpain.pdf Novakovic, B., Fears, T. R., Wexler, L. H., McClure, L. L., Wilson, D. L., McCalla, J. L., et al. (1996). Experiences of cancer in children and adolescents. Cancer Nursing, 19 (1), 54-59. Olson, M. S., Hinds, P. S., Euell, K., Quar gnenti, A., Milligan, M., Foppiano, P., et al. (1998). Peak and nadir experienc es and their consequences described by pediatric oncology nurses. Journal of Pediatric Oncology Nursing, 15(1), 13-24. Panzarella, C., Baggott, C. R., Comeau, M., Duncan, J. M., Groben, V., Woods, D., et al. (2002). Management of disease and treatment-related complications. In C. R. Baggott, K. P. Kelly, D. Fochtman & G. V. Foley (Eds.), Nursing Care of Children and Adolescents with Cancer (3rd ed., pp. 279-318). Philadelphia: Saunders.

PAGE 86

78 Papadatou, D., Bellali, T., Papazoglou, I., & Petraki, D. ( 2002). Greek nurse and physician grief as a result of ca ring for children dying of cancer. Pediatric Nursing, 28 (4), 345-353. Papadatou, D., Martinson, I. M., & Chung, P. M. (2001). Caring for dying children: a comparative study of nurses' experiences in Greece and Hong Kong. Cancer Nursing, 24 (5), 402-412. Petrova, G. G., Todorova, M. T., & Mate va, N. G. (2005). Prerequisites for the occurrence of burnout syndr ome in oncology nurses. Folia Medica, 47 (2), 39-44. Polit, D. F., & Beck, C. T. (2004). Nursing research : principles and methods (7th ed.). Philadelphia: Lippincott Williams & Wilkins. Portenoy, R. K., Thaler, H. T., Kornblith, A. B., Lepore, J. M. Friedlander-Klar, H., Kiyasu, E., et al. (1994). The Memorial Symp tom Assessment Scale: an instrument for the ev aluation of symptom prev alence, characteristics and distress. European Journal of Cancer, 30A (9), 1326-1336. Rankin-Box, D. (1997). Therapies in prac tice: a survey assessing nurses' use of complementary therapies. Complementary Therapies in Nursing and Midwifery, 3 (4), 92-99. Rhodes, V. A., & Watson, P. M. (1987). Symptom distress--the concept: past and present. Seminars in Oncology Nursing, 3 (4), 242-247. Sherman, R. O. (2006). Leading a multi generational nursing workforce: issues, challenges and strategies. Online J Issues Nurs, 11 (2), 3. Soanes, L., Gibson, F., Bayliss, J., & Hannan, J. (2000). Establishing nursing research priorities on a paediatric haematology, oncology, immunology and infectious diseases unit: a Delphi survey. Eur J Oncol Nurs, 4 (2), 108117. Soanes, L., Gibson, F., Hannan, J., & Bay liss, J. (2003). Establishing nursing research priorities on a paediatric haematology, oncology, immunology

PAGE 87

79 and infectious diseases unit: involving doctors and parents. Eur J Oncol Nurs, 7 (2), 110-119. Steen, B., Burghen, E., Hinds, P. S., Sr ivastava, D. K., & Tong, X. (2003). Development and testing of the Role -related Meaning Scale for Staff in Pediatric Oncology. Cancer Nursing, 26 (3), 187-194. Tracy, M. F., Lindquist, R., Savik, K., Watanuki, S., S endelbach, S., Kreitzer, M. J., et al. (2005). Use of complem entary and alternative therapies: a national survey of critical care nurses. American Journal of Critical Care, 14 (5), 404-414; quiz 415-416. Traynor, M., & Wade, B. (1993). The development of a measure of job satisfaction for use in monitoring the mo rale of community nurses in four trusts. Journal of Advanced Nursing, 18 (1), 127-136. U.S. Census Bureau. (2007). Populat ion estimates: Geographic terms and definitions. Retrieved May 19, 2007, from http://www.census.gov/popest/geographic/ Valdres, R. U., Escalante, C., & M anzullo, E. (2001). Fa tigue: a debilitating symptom. Nursing Clinics of North America, 36 (4), 685-694, vi. Wade, B. E. (1993). The job satisfaction of health visitors, district nurses and practice nurses working in areas served by four trusts: year 1. Journal of Advanced Nursing, 18 (6), 992-1004. Williams, P. D., Schmideskamp, J., Ri dder, E. L., & Williams, A. R. (2006). Symptom monitoring and dependent care during cancer treatment in children: Pilot study. Cancer Nursing, 29 (3), 188-197. Wolfe, J., Grier, H. E., Klar N., Levin, S. B., Ellenbog en, J. M., Salem-Schatz, S., et al. (2000). Symptoms and suffering at the end of life in children with cancer. New England Journal of Medicine, 342 (5), 326-333.

PAGE 88

80 Wong, D. L., Hockenberry, M. J. Wilson, D. L., Winkelstein, M. L., & K line, N. E. (Eds.). (2003). Wong's nursing care of infants and children (7 ed.). St. Louis, MO: Mosby. Woodgate, R. L. (2005). A di fferent way of being: adole scents' experiences with cancer. Cancer Nursing, 28 (1), 8-15. Woodgate, R. L. (2006). Life is never the same: childhood cancer narratives. European Journal of Cancer Care (English Language Edition), 15 (1), 8-18. Woodgate, R. L., & Degner, L. F. (2004). Cancer symptom transition periods of children and families. Journal of Advanced Nursing, 46 (4), 358-368. Woodgate, R. L., Degner, L. F., & Yanofsky, R. (2003). A different perspective to approaching cancer sym ptoms in children. Journal of Pain and Symptom Management, 26 (3), 800-817.

PAGE 89

81 Appendices

PAGE 90

82 Appendix A: Nurses Distress and Intervent ions for Symptoms Survey (NDISS) 1. In the past month, have any of your patients experienced PAIN? No SKIP to #3 Yes 2. How much did it distress or bother you that your patients had PAIN? (Please select one response) Not At All (0) A Little Bit (1) Somewhat (2) Quite a Bit (3) Very Much (4) 3. Which of the following do you normally use to help treat PAIN? Include what you personally administer and what you arrange for someone else to administer. (Please check all that apply) Active listening Adjust nighttime sleep regimen Anticipatory guidance (explain what is happening) Anxiety-reducing medications Art therapy Assist with physical needs Build trust Counseling Decision-making support Deep breathing Distraction Education Emotional support Encourage family involvement Family support Humor Imagery Meditation (continued in next column) Mouth care/hygiene Music therapy Mutual goal-setting Nausea-reducing medications Nutrition Pain-reducing medications Pet therapy Play therapy Positioning Prayer Presence Psychosocial support for patient Reduced sleep interruptions Relaxation Sleep-inducing medications Spiritual support Stress management Other: ____________________ Other: ____________________ Other: 4. How effective do you feel you are at ma naging PAIN using these interventions? (Please select one response) Not At All (0) A Little Bit (1) Somewhat (2) Quite a Bit (3) Very Much (4)

PAGE 91

83 Appendix A: (Continued) 5. In the past month, have any of your patients experienced NAUSEA/VOMITING? No SKIP to #7 Yes 6. How much did it distress or bother you that your patients had NAUSEA/VOMITING? (Please select one response) Not At All (0) A Little Bit (1) Somewhat (2) Quite a Bit (3) Very Much (4) 7. Which of the following do you normally use to help treat NAUSEA/VOMITING? Include what you personally administer and what you arrange for someone else to administer. (Please check all that apply) Active listening Adjust nighttime sleep regimen Anticipatory guidance (explain what is happening) Anxiety-reducing medications Art therapy Assist with physical needs Build trust Counseling Decision-making support Deep breathing Distraction Education Emotional support Encourage family involvement Family support Humor Imagery Meditation (continued in next column) Mouth care/hygiene Music therapy Mutual goal-setting Nausea-reducing medications Nutrition Pain-reducing medications Pet therapy Play therapy Positioning Prayer Presence Psychosocial support for patient Reduced sleep interruptions Relaxation Sleep-inducing medications Spiritual support Stress management Other: ____________________ Other: ____________________ Other: 8. How effective do you feel you are at managing NAUSEA/VOMITING using these interventions? (Please select one response) Not At All (0) A Little Bit (1) Somewhat (2) Quite a Bit (3) Very Much (4)

PAGE 92

84 Appendix A: (Continued) 9. In the past month, have any of your patients experienced MOUTH SORES? No SKIP to #11 Yes 10. How much did it distress or bother you that your patients had MOUTH SORES? (Please select one response) Not At All (0) A Little Bit (1) Somewhat (2) Quite a Bit (3) Very Much (4) 11. Which of the following do you normally use to help treat MOUTH SORES? Include what you personally administer and what you arrange for someone else to administer. (Please check all that apply) Active listening Adjust nighttime sleep regimen Anticipatory guidance (explain what is happening) Anxiety-reducing medications Art therapy Assist with physical needs Build trust Counseling Decision-making support Deep breathing Distraction Education Emotional support Encourage family involvement Family support Humor Imagery Meditation (continued in next column) Mouth care/hygiene Music therapy Mutual goal-setting Nausea-reducing medications Nutrition Pain-reducing medications Pet therapy Play therapy Positioning Prayer Presence Psychosocial support for patient Reduced sleep interruptions Relaxation Sleep-inducing medications Spiritual support Stress management Other: ____________________ Other: ____________________ Other: ____________________ 12. How effective do you feel you are at managing MOUTH SORES using these interventions? (Please select one response) Not At All (0) A Little Bit (1) Somewhat (2) Quite a Bit (3) Very Much (4)

PAGE 93

85 Appendix A: (Continued) 13. In the past month, have any of your patients experienced TROUBLE SLEEPING? No SKIP to #15 Yes 14. How much did it distress or bother you that your patients had TROUBLE SLEEPING? (Please select one response) Not At All (0) A Little Bit (1) Somewhat (2) Quite a Bit (3) Very Much (4) 15. Which of the following do you norma lly use to help treat TROUBLE SLEEPING? Include what you personally administer and what you arrange for someone else to administer. (Please check all that apply) Active listening Adjust nighttime sleep regimen Anticipatory guidance (explain what is happening) Anxiety-reducing medications Art therapy Assist with physical needs Build trust Counseling Decision-making support Deep breathing Distraction Education Emotional support Encourage family involvement Family support Humor Imagery Meditation (continued in next column) Mouth care/hygiene Music therapy Mutual goal-setting Nausea-reducing medications Nutrition Pain-reducing medications Pet therapy Play therapy Positioning Prayer Presence Psychosocial support for patient Reduced sleep interruptions Relaxation Sleep-inducing medications Spiritual support Stress management Other: ____________________ Other: ____________________ Other: 16. How effective do you feel you are at mana ging TROUBLE SLEEPING using these interventions? (Please select one response) Not At All (0) A Little Bit (1) Somewhat (2) Quite a Bit (3) Very Much (4)

PAGE 94

86 Appendix A: (Continued) 17. In the past month, have any of your patients experienced FATIGUE? No SKIP to #19 Yes 18. How much did it distress or bother you that your patients had FATIGUE? (Please select one response) Not At All (0) A Little Bit (1) Somewhat (2) Quite a Bit (3) Very Much (4) 19. Which of the following do you normally use to help treat FATIGUE? Include what you personally administer and what you arrange for someone else to administer. (Please check all that apply) Active listening Adjust nighttime sleep regimen Anticipatory guidance (explain what is happening) Anxiety-reducing medications Art therapy Assist with physical needs Build trust Counseling Decision-making support Deep breathing Distraction Education Emotional support Encourage family involvement Family support Humor Imagery Meditation (continued in next column) Mouth care/hygiene Music therapy Mutual goal-setting Nausea-reducing medications Nutrition Pain-reducing medications Pet therapy Play therapy Positioning Prayer Presence Psychosocial support for patient Reduced sleep interruptions Relaxation Sleep-inducing medications Spiritual support Stress management Other: ____________________ Other: ____________________ Other: 20. How effective do you feel you are at managing FATIGUE using these interventions? (Please select one response) Not At All (0) A Little Bit (1) Somewhat (2) Quite a Bit (3) Very Much (4)

PAGE 95

87 Appendix A: (Continued) 21. In the past month, have any of your patients experienced WORRY? No SKIP to #23 Yes 22. How much did it distress or bother you that your patients had WORRY? (Please select one response) Not At All (0) A Little Bit (1) Somewhat (2) Quite a Bit (3) Very Much (4) 23. Which of the following do you normally use to help treat WORRY? Include what you personally administer and what you arrange for someone else to administer. (Please check all that apply) Active listening Adjust nighttime sleep regimen Anticipatory guidance (explain what is happening) Anxiety-reducing medications Art therapy Assist with physical needs Build trust Counseling Decision-making support Deep breathing Distraction Education Emotional support Encourage family involvement Family support Humor Imagery Meditation (continued in next column) Mouth care/hygiene Music therapy Mutual goal-setting Nausea-reducing medications Nutrition Pain-reducing medications Pet therapy Play therapy Positioning Prayer Presence Psychosocial support for patient Reduced sleep interruptions Relaxation Sleep-inducing medications Spiritual support Stress management Other: ____________________ Other: ____________________ Other: 24. How effective do you feel you are at managing WORRY using these interventions? (Please select one response) Not At All (0) A Little Bit (1) Somewhat (2) Quite a Bit (3) Very Much (4)

PAGE 96

88 Appendix A: (Continued) 25. In the past month, have any of your patients experienced HAIR LOSS? No SKIP to #27 Yes 26. How much did it distress or bother you that your patients had HAIR LOSS? (Please select one response) Not At All (0) A Little Bit (1) Somewhat (2) Quite a Bit (3) Very Much (4) 27. Which of the following do you normally use to help treat HAIR LOSS? Include what you personally administer and what you arrange for someone else to administer. (Please check all that apply) Active listening Adjust nighttime sleep regimen Anticipatory guidance (explain what is happening) Anxiety-reducing medications Art therapy Assist with physical needs Build trust Counseling Decision-making support Deep breathing Distraction Education Emotional support Encourage family involvement Family support Humor Imagery Meditation (continued in next column) Mouth care/hygiene Music therapy Mutual goal-setting Nausea-reducing medications Nutrition Pain-reducing medications Pet therapy Play therapy Positioning Prayer Presence Psychosocial support for patient Reduced sleep interruptions Relaxation Sleep-inducing medications Spiritual support Stress management Other: ____________________ Other: ____________________ Other: 28. How effective do you feel you are at managing HAIR LOSS using these interventions? (Please select one response) Not At All (0) A Little Bit (1) Somewhat (2) Quite a Bit (3) Very Much (4)

PAGE 97

89 Appendix B: Nurses’ Ranked Use of Interventions in Phase Three Nursing intervention OverallPainN/V Mouth sores Trouble sleeping Fatigue Worry Hair loss Emotional support 1 3 4 7 6 2 2 1 Encourage family involvement 2 5 5 4 7 1 3 7 Active listening 3 4 9 10 8 8 1 2 Family support 4 13 8 8 9 6 4 3 Education 5 12 7 5 12 9 7 4 Anticipatory guidance 6 8 6 6 14 11 8 5 Psychosocial support for patient 7 15 13 12 10 10 5 6 Anxiety-reducing medications 8 9 2 14 4 17 9 22 Build trust 9 10 15 13 15 13 6 9 Distraction 10 2 3 9 20 4 15 19 Relaxation 11 19 11 16 5 7 12 20 Assist with physical needs 12 11 12 11 16 3 20 14 Pain-reducing medications 13 1 24 2 11 18 28 34 Nausea-reducing medications 14 7 1 15 19 26 30 32 Reduced sleep interruptions 15 21 20 19 3 4 25 8 Humor 16 16 18 21 29 22 14 8 Presence 17 20 17 17 18 20 13 12 Decision-making support 18 22 22 20 22 16 10 13 Play therapy 19 18 21 18 31 21 16 11 Mouth care/ hygiene 20 17 16 1 33 36 35 31 Sleep-inducing medications 21 23 19 23 2 14 29 35 Nutrition 22 24 14 3 30 12 34 27 Deep breathing 23 14 10 5 17 28 21 29 Counseling 24 28 26 24 23 19 11 10 Positioning 25 6 23 30 13 23 33 33 Adjust nighttime sleep regimen 26 35 34 35 1 5 32 36 Mutual goal-setting 27 25 25 22 26 15 18 21 Stress management 28 32 28 26 21 25 19 15 Spiritual support 29 29 33 32 28 29 17 17 Music therapy 30 26 29 27 24 30 23 26 Art therapy 31 27 31 29 35 32 22 18 Imagery 32 30 27 34 25 33 27 25 Prayer 33 33 35 36 32 35 24 23 Pet therapy 34 31 32 31 36 34 26 24 Meditation 35 34 30 33 27 31 31 30 Other (not listed) 36 36 36 28 34 27 36 16 Note. Results in rank order. N/V = Nausea/Vomiting.

PAGE 98

90 About the Author Jennifer I. Rheingans received a Bachel or’s Degree in Nursing from the University of Florida in 1998. She began pr acticing as a Registered Nurse (RN) in 1998 with a specialty in pediatrics and then in pediatric oncology. In 2004 she entered the BS to PhD in Nursing program at the University of South Florida, and received her Master’s of Science in Nurs ing with a focus in nursing education in Summer 2006. She has worked at Saraso ta Memorial Hosp ital throughout her graduate studies and is currently a Clin ical Nurse Researcher supporting Evidence-Based Practice for nurses at her hospital. Mrs. Rheingans has maintained her ce rtification as a pediatric oncology nurse, and has strongly pursued holistic nursing and patientand family-centered care. She has presented posters and published papers about complementary therapies in pediatric oncology. She is preparing to take the Holistic Nursing certification exam shortl y after doctoral graduation.